Merge branch 'main' into perf/ui-table

# Conflicts:
#	web/default/src/features/channels/components/channels-table.tsx

.env.example

@@ -56,6 +56,8 @@
 # 对话超时设置
 # 所有请求超时时间，单位秒，默认为0，表示不限制
 # RELAY_TIMEOUT=0
+# Relay HTTP 客户端空闲连接超时时间，单位秒，默认跟随 Go 标准库，设置为0表示不限制
+# RELAY_IDLE_CONN_TIMEOUT=90
 # 流模式无响应超时时间，单位秒，如果出现空补全可以尝试改为更大值
 # STREAMING_TIMEOUT=300
 

.github/ISSUE_TEMPLATE/bug_report.md

@@ -11,6 +11,8 @@ assignees: ''
 
 - 文档：https://docs.newapi.ai/
 - 使用问题先看或先问：https://deepwiki.com/QuantumNous/new-api
+- 开启透传后的转发相关反馈不接受 issue；透传模式会直接转发请求，请自行确认上游行为。
+- 不接受 coding plan、逆向渠道等技术支持类 issue。
 - 警告：删除本模板、删除小节标题或随意清空内容的 issue，可能会被直接关闭；重复恶意提交者可能会被 block。
 
 **您当前的 newapi 版本**
@@ -20,13 +22,18 @@ assignees: ''
 **提交确认**
 
 [//]: # (方框内删除已有的空格，填 x 号)
-+ [ ] 我已确认目前没有类似 issue
-+ [ ] 我已完整查看过文档 https://docs.newapi.ai/ 和项目 README，尤其是常见问题部分
-+ [ ] 我未删除此模板中的任何引导内容或小节标题，并会按要求完整填写
-+ [ ] 我理解项目维护者精力有限，不遵循模板要求的 issue 可能会被无视或直接关闭
+- [ ] **非重复 issue:** 我已搜索现有 [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue)，确认目前没有类似 issue。
+- [ ] **提交前必读:** 我已完整阅读上方“提交前必读”，并已查看文档 https://docs.newapi.ai/、项目 README 且向 AI 提问，确认这不是使用、配置或接入类问题。
+- [ ] **模板完整:** 我未删除此模板中的任何引导内容或小节标题，并会按要求完整填写。
+- [ ] **维护成本:** 我理解项目维护者精力有限，不遵循模板要求的 issue 可能会被无视或直接关闭。
 
 **问题描述**
 
+请尽可能说明问题现象、影响范围，以及你判断它是程序问题而不是上游行为或使用问题的依据。
+
+- 转发问题请尽可能说明渠道类型、转换格式、上游原生支持依据和服务端日志。
+- 计费问题请尽可能附请求返回的 `usage` 示例。
+
 **复现步骤**
 
 **预期结果**

.github/ISSUE_TEMPLATE/bug_report_en.md

@@ -11,6 +11,8 @@ assignees: ''
 
 - Docs: https://docs.newapi.ai/
 - Usage questions first: https://deepwiki.com/QuantumNous/new-api
+- Issues about forwarding behavior after enabling pass-through mode are not accepted; pass-through mode forwards requests directly, so please verify upstream behavior yourself.
+- Technical support requests such as coding plans or reverse-engineering channels are not accepted as issues.
 - Warning: issues with this template removed, section headings deleted, or content cleared may be closed directly. Repeated abusive submissions may result in a block.
 
 **Your current newapi version**
@@ -20,13 +22,18 @@ Please fill this in, for example: `v1.0.0`
 **Submission Checks**
 
 [//]: # (Remove the space in the box and fill with an x)
-+ [ ] I have confirmed there are no similar issues
-+ [ ] I have thoroughly read the docs at https://docs.newapi.ai/ and the project README, especially the FAQ section
-+ [ ] I have not removed any guidance or section headings from this template and will complete it as requested
-+ [ ] I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly
+- [ ] **Non-duplicate issue:** I have searched existing [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue) and confirmed there are no similar issues.
+- [ ] **Read this first:** I have fully read the section above, reviewed the docs at https://docs.newapi.ai/ and the project README, and asked AI first, confirming this is not a usage, configuration, or integration question.
+- [ ] **Template intact:** I have not removed any guidance or section headings from this template and will complete it as requested.
+- [ ] **Maintainer time:** I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly.
 
 **Issue Description**
 
+Describe the symptom, impact scope, and why you believe this is an application issue rather than upstream behavior or a usage question with as much detail as possible.
+
+- For forwarding issues, include the channel type, conversion format, upstream native-support evidence, and server logs when possible.
+- For billing issues, include an example of the returned `usage` when possible.
+
 **Steps to Reproduce**
 
 **Expected Result**

.github/ISSUE_TEMPLATE/feature_request.md

@@ -11,6 +11,8 @@ assignees: ''
 
 - 文档：https://docs.newapi.ai/
 - 使用问题先看或先问：https://deepwiki.com/QuantumNous/new-api
+- 开启透传后的转发相关反馈不接受 issue；透传模式会直接转发请求，请自行确认上游行为。
+- 不接受 coding plan、逆向渠道等技术支持类 issue。
 - 警告：删除本模板、删除小节标题或随意清空内容的 issue，可能会被直接关闭；重复恶意提交者可能会被 block。
 
 **您当前的 newapi 版本**
@@ -20,10 +22,10 @@ assignees: ''
 **提交确认**
 
 [//]: # (方框内删除已有的空格，填 x 号)
-+ [ ] 我已确认目前没有类似 issue
-+ [ ] 我已完整查看过文档 https://docs.newapi.ai/ 和项目 README，已确定现有版本无法满足需求
-+ [ ] 我未删除此模板中的任何引导内容或小节标题，并会按要求完整填写
-+ [ ] 我理解项目维护者精力有限，不遵循模板要求的 issue 可能会被无视或直接关闭
+- [ ] **非重复 issue:** 我已搜索现有 [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue)，确认目前没有类似 issue。
+- [ ] **提交前必读:** 我已完整阅读上方“提交前必读”，并已查看文档 https://docs.newapi.ai/、项目 README 且向 AI 提问，确认这不是使用、配置或接入类问题，且现有版本无法满足需求。
+- [ ] **模板完整:** 我未删除此模板中的任何引导内容或小节标题，并会按要求完整填写。
+- [ ] **维护成本:** 我理解项目维护者精力有限，不遵循模板要求的 issue 可能会被无视或直接关闭。
 
 **功能描述**
 

.github/ISSUE_TEMPLATE/feature_request_en.md

@@ -11,6 +11,8 @@ assignees: ''
 
 - Docs: https://docs.newapi.ai/
 - Usage questions first: https://deepwiki.com/QuantumNous/new-api
+- Issues about forwarding behavior after enabling pass-through mode are not accepted; pass-through mode forwards requests directly, so please verify upstream behavior yourself.
+- Technical support requests such as coding plans or reverse-engineering channels are not accepted as issues.
 - Warning: issues with this template removed, section headings deleted, or content cleared may be closed directly. Repeated abusive submissions may result in a block.
 
 **Your current newapi version**
@@ -20,10 +22,10 @@ Please fill this in, for example: `v1.0.0`
 **Submission Checks**
 
 [//]: # (Remove the space in the box and fill with an x)
-+ [ ] I have confirmed there are no similar issues
-+ [ ] I have thoroughly read the docs at https://docs.newapi.ai/ and the project README, and confirmed the current version cannot meet my needs
-+ [ ] I have not removed any guidance or section headings from this template and will complete it as requested
-+ [ ] I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly
+- [ ] **Non-duplicate issue:** I have searched existing [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue) and confirmed there are no similar issues.
+- [ ] **Read this first:** I have fully read the section above, reviewed the docs at https://docs.newapi.ai/ and the project README, and asked AI first, confirming this is not a usage, configuration, or integration question, and that the current version cannot meet my needs.
+- [ ] **Template intact:** I have not removed any guidance or section headings from this template and will complete it as requested.
+- [ ] **Maintainer time:** I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly.
 
 **Feature Description**
 

.github/workflows/release.yml

@@ -33,16 +33,18 @@ jobs:
         env:
           CI: ""
         run: |
-          cd web/default
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd default
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Build Frontend (classic)
         env:
           CI: ""
         run: |
-          cd web/classic
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd classic
           VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Set up Go
@@ -91,16 +93,18 @@ jobs:
           CI: ""
           NODE_OPTIONS: "--max-old-space-size=4096"
         run: |
-          cd web/default
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd default
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Build Frontend (classic)
         env:
           CI: ""
         run: |
-          cd web/classic
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd classic
           VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Set up Go
@@ -146,16 +150,18 @@ jobs:
         env:
           CI: ""
         run: |
-          cd web/default
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd default
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Build Frontend (classic)
         env:
           CI: ""
         run: |
-          cd web/classic
-          bun install
+          cd web
+          bun install --frozen-lockfile
+          cd classic
           VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ../..
       - name: Set up Go

.gitignore

@@ -35,3 +35,4 @@ data/
 .test
 token_estimator_test.go
 skills-lock.json
+.playwright-mcp

Dockerfile

@@ -1,22 +1,24 @@
 FROM oven/bun:1@sha256:0733e50325078969732ebe3b15ce4c4be5082f18c4ac1a0f0ca4839c2e4e42a7 AS builder
 
-WORKDIR /build
-COPY web/default/package.json .
-COPY web/default/bun.lock .
-RUN bun install
-COPY ./web/default .
-COPY ./VERSION .
-RUN DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
+WORKDIR /build/web
+COPY web/package.json web/bun.lock ./
+COPY web/default/package.json ./default/package.json
+COPY web/classic/package.json ./classic/package.json
+RUN bun install --frozen-lockfile
+COPY ./web/default ./default
+COPY ./VERSION /build/VERSION
+RUN cd default && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat /build/VERSION) bun run build
 
 FROM oven/bun:1@sha256:0733e50325078969732ebe3b15ce4c4be5082f18c4ac1a0f0ca4839c2e4e42a7 AS builder-classic
 
-WORKDIR /build
-COPY web/classic/package.json .
-COPY web/classic/bun.lock .
-RUN bun install
-COPY ./web/classic .
-COPY ./VERSION .
-RUN VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
+WORKDIR /build/web
+COPY web/package.json web/bun.lock ./
+COPY web/default/package.json ./default/package.json
+COPY web/classic/package.json ./classic/package.json
+RUN bun install --frozen-lockfile
+COPY ./web/classic ./classic
+COPY ./VERSION /build/VERSION
+RUN cd classic && VITE_REACT_APP_VERSION=$(cat /build/VERSION) bun run build
 
 FROM golang:1.26.1-alpine@sha256:2389ebfa5b7f43eeafbd6be0c3700cc46690ef842ad962f6c5bd6be49ed82039 AS builder2
 ENV GO111MODULE=on CGO_ENABLED=0
@@ -32,8 +34,8 @@ ADD go.mod go.sum ./
 RUN go mod download
 
 COPY . .
-COPY --from=builder /build/dist ./web/default/dist
-COPY --from=builder-classic /build/dist ./web/classic/dist
+COPY --from=builder /build/web/default/dist ./web/default/dist
+COPY --from=builder-classic /build/web/classic/dist ./web/classic/dist
 RUN go build -ldflags "-s -w -X 'github.com/QuantumNous/new-api/common.Version=$(cat VERSION)'" -o new-api
 
 FROM debian:bookworm-slim@sha256:f06537653ac770703bc45b4b113475bd402f451e85223f0f2837acbf89ab020a

README.md

@@ -316,6 +316,7 @@ docker run --name new-api -d --restart always \
 | `CRYPTO_SECRET` | Encryption secret (required for Redis) | - |
 | `SQL_DSN` | Database connection string | - |
 | `REDIS_CONN_STRING` | Redis connection string | - |
+| `RELAY_IDLE_CONN_TIMEOUT` | Idle keep-alive timeout for relay HTTP clients, seconds. Defaults to Go standard library behavior; set `0` to disable | `90` |
 | `STREAMING_TIMEOUT` | Streaming timeout (seconds) | `300` |
 | `STREAM_SCANNER_MAX_BUFFER_MB` | Max per-line buffer (MB) for the stream scanner; increase when upstream sends huge image/base64 payloads | `64` |
 | `MAX_REQUEST_BODY_MB` | Max request body size (MB, counted **after decompression**; prevents huge requests/zip bombs from exhausting memory). Exceeding it returns `413` | `32` |

common/constants.go

@@ -170,6 +170,7 @@ var BatchUpdateInterval int
 
 var RelayTimeout int // unit is second
 
+var RelayIdleConnTimeout int // unit is second
 var RelayMaxIdleConns int
 var RelayMaxIdleConnsPerHost int
 

common/custom-event.go

@@ -37,7 +37,7 @@ func checkWriter(writer io.Writer) stringWriter {
 // W3C Working Draft 29 October 2009
 // http://www.w3.org/TR/2009/WD-eventsource-20091029/
 
-var contentType = []string{"text/event-stream"}
+var writeContentType = []string{"text/event-stream"}
 var noCache = []string{"no-cache"}
 
 var fieldReplacer = strings.NewReplacer(
@@ -79,7 +79,7 @@ func (r CustomEvent) WriteContentType(w http.ResponseWriter) {
 	r.Mutex.Lock()
 	defer r.Mutex.Unlock()
 	header := w.Header()
-	header["Content-Type"] = contentType
+	header["Content-Type"] = writeContentType
 
 	if _, exist := header["Cache-Control"]; !exist {
 		header["Cache-Control"] = noCache

common/gin.go

@@ -110,11 +110,29 @@ func UnmarshalBodyReusable(c *gin.Context, v any) error {
 	if err != nil {
 		return err
 	}
+	contentType := c.Request.Header.Get("Content-Type")
+
+	// disk-backed JSON: stream-decode directly from the file to avoid
+	// materializing the entire payload back into a transient []byte
+	// (diskStorage.Bytes() would ReadFull the whole file into the heap).
+	if storage.IsDisk() && strings.HasPrefix(contentType, "application/json") {
+		if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
+			return seekErr
+		}
+		if err := DecodeJson(storage, v); err != nil {
+			return err
+		}
+		if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
+			return seekErr
+		}
+		c.Request.Body = io.NopCloser(storage)
+		return nil
+	}
+
 	requestBody, err := storage.Bytes()
 	if err != nil {
 		return err
 	}
-	contentType := c.Request.Header.Get("Content-Type")
 	if strings.HasPrefix(contentType, "application/json") {
 		err = Unmarshal(requestBody, v)
 	} else if strings.Contains(contentType, gin.MIMEPOSTForm) {

common/init.go

@@ -102,6 +102,7 @@ func InitEnv() {
 	SyncFrequency = GetEnvOrDefault("SYNC_FREQUENCY", 60)
 	BatchUpdateInterval = GetEnvOrDefault("BATCH_UPDATE_INTERVAL", 5)
 	RelayTimeout = GetEnvOrDefault("RELAY_TIMEOUT", 0)
+	RelayIdleConnTimeout = GetEnvOrDefault("RELAY_IDLE_CONN_TIMEOUT", 90)
 	RelayMaxIdleConns = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS", 500)
 	RelayMaxIdleConnsPerHost = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS_PER_HOST", 100)
 
@@ -111,11 +112,11 @@ func InitEnv() {
 
 	// Initialize rate limit variables
 	GlobalApiRateLimitEnable = GetEnvOrDefaultBool("GLOBAL_API_RATE_LIMIT_ENABLE", true)
-	GlobalApiRateLimitNum = GetEnvOrDefault("GLOBAL_API_RATE_LIMIT", 180)
+	GlobalApiRateLimitNum = GetEnvOrDefault("GLOBAL_API_RATE_LIMIT", 360)
 	GlobalApiRateLimitDuration = int64(GetEnvOrDefault("GLOBAL_API_RATE_LIMIT_DURATION", 180))
 
 	GlobalWebRateLimitEnable = GetEnvOrDefaultBool("GLOBAL_WEB_RATE_LIMIT_ENABLE", true)
-	GlobalWebRateLimitNum = GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT", 60)
+	GlobalWebRateLimitNum = GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT", 120)
 	GlobalWebRateLimitDuration = int64(GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT_DURATION", 180))
 
 	CriticalRateLimitEnable = GetEnvOrDefaultBool("CRITICAL_RATE_LIMIT_ENABLE", true)
@@ -135,6 +136,7 @@ func initConstantEnv() {
 	constant.StreamScannerMaxBufferMB = GetEnvOrDefault("STREAM_SCANNER_MAX_BUFFER_MB", 128)
 	// MaxRequestBodyMB 请求体最大大小（解压后），用于防止超大请求/zip bomb导致内存暴涨
 	constant.MaxRequestBodyMB = GetEnvOrDefault("MAX_REQUEST_BODY_MB", 128)
+	constant.AnonymousRequestBodyLimitKB = GetEnvOrDefault("ANONYMOUS_REQUEST_BODY_LIMIT_KB", 512)
 	// ForceStreamOption 覆盖请求参数，强制返回usage信息
 	constant.ForceStreamOption = GetEnvOrDefaultBool("FORCE_STREAM_OPTION", true)
 	constant.CountToken = GetEnvOrDefaultBool("CountToken", true)

common/request_body_limit.go

@@ -0,0 +1,13 @@
+package common
+
+import "github.com/QuantumNous/new-api/constant"
+
+const defaultAnonymousRequestBodyLimitKB = 512
+
+func GetAnonymousRequestBodyLimitBytes() int64 {
+	limitKB := constant.AnonymousRequestBodyLimitKB
+	if limitKB < 0 {
+		limitKB = defaultAnonymousRequestBodyLimitKB
+	}
+	return int64(limitKB) << 10
+}

common/str.go

@@ -3,6 +3,7 @@ package common
 import (
 	"encoding/base64"
 	"encoding/json"
+	"fmt"
 	"net/url"
 	"regexp"
 	"strconv"
@@ -20,6 +21,16 @@ var (
 	maskApiKeyPattern = regexp.MustCompile(`(['"]?)api_key:([^\s'"]+)(['"]?)`)
 )
 
+const LocalLogContentLimit = 2048
+
+// LocalLogPreview limits log-only content unless debug logging is enabled.
+func LocalLogPreview(content string) string {
+	if DebugEnabled || len(content) <= LocalLogContentLimit {
+		return content
+	}
+	return fmt.Sprintf("%s... [truncated, original_length=%d, limit=%d]", content[:LocalLogContentLimit], len(content), LocalLogContentLimit)
+}
+
 func GetStringIfEmpty(str string, defaultValue string) string {
 	if str == "" {
 		return defaultValue

constant/env.go

@@ -10,6 +10,7 @@ var GetMediaToken bool
 var GetMediaTokenNotStream bool
 var UpdateTask bool
 var MaxRequestBodyMB int
+var AnonymousRequestBodyLimitKB int
 var AzureDefaultAPIVersion string
 var NotifyLimitCount int
 var NotificationLimitDurationMinute int

controller/channel-test.go

@@ -57,7 +57,24 @@ func normalizeChannelTestEndpoint(channel *model.Channel, modelName, endpointTyp
 	return normalized
 }
 
-func testChannel(channel *model.Channel, testModel string, endpointType string, isStream bool) testResult {
+func resolveChannelTestUserID(c *gin.Context) (int, error) {
+	if c != nil {
+		if userID := c.GetInt("id"); userID > 0 {
+			return userID, nil
+		}
+	}
+
+	var rootUser model.User
+	if err := model.DB.Select("id").Where("role = ?", common.RoleRootUser).First(&rootUser).Error; err != nil {
+		return 0, fmt.Errorf("failed to resolve channel test user: %w", err)
+	}
+	if rootUser.Id == 0 {
+		return 0, errors.New("failed to resolve channel test user")
+	}
+	return rootUser.Id, nil
+}
+
+func testChannel(channel *model.Channel, testUserID int, testModel string, endpointType string, isStream bool) testResult {
 	tik := time.Now()
 	var unsupportedTestChannelTypes = []int{
 		constant.ChannelTypeMidjourney,
@@ -143,7 +160,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
 		Header: make(http.Header),
 	}
 
-	cache, err := model.GetUserCache(1)
+	cache, err := model.GetUserCache(testUserID)
 	if err != nil {
 		return testResult{
 			localErr:    err,
@@ -151,13 +168,13 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
 		}
 	}
 	cache.WriteContext(c)
-	c.Set("id", 1)
+	c.Set("id", testUserID)
 
 	//c.Request.Header.Set("Authorization", "Bearer "+channel.Key)
 	c.Request.Header.Set("Content-Type", "application/json")
 	c.Set("channel", channel.Type)
 	c.Set("base_url", channel.GetBaseURL())
-	group, _ := model.GetUserGroup(1, false)
+	group, _ := model.GetUserGroup(testUserID, false)
 	c.Set("group", group)
 
 	newAPIError := middleware.SetupContextForSelectedChannel(c, channel, testModel)
@@ -484,7 +501,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
 	milliseconds := tok.Sub(tik).Milliseconds()
 	consumedTime := float64(milliseconds) / 1000.0
 	other := buildTestLogOther(c, info, priceData, usage, tieredResult)
-	model.RecordConsumeLog(c, 1, model.RecordConsumeLogParams{
+	model.RecordConsumeLog(c, testUserID, model.RecordConsumeLogParams{
 		ChannelId:        channel.Id,
 		PromptTokens:     usage.PromptTokens,
 		CompletionTokens: usage.CompletionTokens,
@@ -797,7 +814,7 @@ func buildTestRequest(model string, endpointType string, channel *model.Channel,
 		testRequest.StreamOptions = &dto.StreamOptions{IncludeUsage: true}
 	}
 
-	if strings.HasPrefix(model, "o") {
+	if dto.IsOpenAIReasoningOModel(model) {
 		testRequest.MaxCompletionTokens = lo.ToPtr(uint(16))
 	} else if strings.Contains(model, "thinking") {
 		if !strings.Contains(model, "claude") {
@@ -834,8 +851,13 @@ func TestChannel(c *gin.Context) {
 	testModel := c.Query("model")
 	endpointType := c.Query("endpoint_type")
 	isStream, _ := strconv.ParseBool(c.Query("stream"))
+	testUserID, err := resolveChannelTestUserID(c)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
 	tik := time.Now()
-	result := testChannel(channel, testModel, endpointType, isStream)
+	result := testChannel(channel, testUserID, testModel, endpointType, isStream)
 	if result.localErr != nil {
 		resp := gin.H{
 			"success": false,
@@ -872,6 +894,10 @@ var testAllChannelsLock sync.Mutex
 var testAllChannelsRunning bool = false
 
 func testAllChannels(notify bool) error {
+	testUserID, err := resolveChannelTestUserID(nil)
+	if err != nil {
+		return err
+	}
 
 	testAllChannelsLock.Lock()
 	if testAllChannelsRunning {
@@ -902,7 +928,7 @@ func testAllChannels(notify bool) error {
 			}
 			isChannelEnabled := channel.Status == common.ChannelStatusEnabled
 			tik := time.Now()
-			result := testChannel(channel, "", "", shouldUseStreamForAutomaticChannelTest(channel))
+			result := testChannel(channel, testUserID, "", "", shouldUseStreamForAutomaticChannelTest(channel))
 			tok := time.Now()
 			milliseconds := tok.Sub(tik).Milliseconds()
 

controller/channel.go

@@ -19,6 +19,7 @@ import (
 	"github.com/QuantumNous/new-api/service"
 
 	"github.com/gin-gonic/gin"
+	"gorm.io/gorm"
 )
 
 type OpenAIModel struct {
@@ -68,12 +69,33 @@ func clearChannelInfo(channel *model.Channel) {
 	}
 }
 
+func applyChannelStatusFilter(query *gorm.DB, statusFilter int) *gorm.DB {
+	if statusFilter == common.ChannelStatusEnabled {
+		return query.Where("status = ?", common.ChannelStatusEnabled)
+	}
+	if statusFilter == 0 {
+		return query.Where("status != ?", common.ChannelStatusEnabled)
+	}
+	return query
+}
+
+func buildChannelListQuery(group string, statusFilter int, typeFilter int) *gorm.DB {
+	query := model.DB.Model(&model.Channel{})
+	query = model.ApplyChannelGroupFilter(query, group)
+	query = applyChannelStatusFilter(query, statusFilter)
+	if typeFilter >= 0 {
+		query = query.Where("type = ?", typeFilter)
+	}
+	return query
+}
+
 func GetAllChannels(c *gin.Context) {
 	pageInfo := common.GetPageQuery(c)
 	channelData := make([]*model.Channel, 0)
 	idSort, _ := strconv.ParseBool(c.Query("id_sort"))
 	sortOptions := model.NewChannelSortOptions(c.Query("sort_by"), c.Query("sort_order"), idSort)
 	enableTagMode, _ := strconv.ParseBool(c.Query("tag_mode"))
+	groupFilter := model.NormalizeChannelGroupFilter(c.Query("group"))
 	statusParam := c.Query("status")
 	// statusFilter: -1 all, 1 enabled, 0 disabled (include auto & manual)
 	statusFilter := parseStatusFilter(statusParam)
@@ -89,50 +111,45 @@ func GetAllChannels(c *gin.Context) {
 	var total int64
 
 	if enableTagMode {
-		tags, err := model.GetPaginatedTags(pageInfo.GetStartIdx(), pageInfo.GetPageSize())
+		tags, err := model.GetPaginatedChannelTags(buildChannelListQuery(groupFilter, statusFilter, typeFilter), pageInfo.GetStartIdx(), pageInfo.GetPageSize())
 		if err != nil {
 			common.SysError("failed to get paginated tags: " + err.Error())
 			c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取标签失败，请稍后重试"})
 			return
 		}
+		total, err = model.CountChannelTags(buildChannelListQuery(groupFilter, statusFilter, typeFilter))
+		if err != nil {
+			common.SysError("failed to count tags: " + err.Error())
+			c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取标签数量失败，请稍后重试"})
+			return
+		}
 		for _, tag := range tags {
 			if tag == nil || *tag == "" {
 				continue
 			}
-			tagChannels, err := model.GetChannelsByTag(*tag, idSort, false, sortOptions)
+			var tagChannels []*model.Channel
+			err := sortOptions.Apply(buildChannelListQuery(groupFilter, statusFilter, typeFilter).Where("tag = ?", *tag)).
+				Omit("key").
+				Find(&tagChannels).Error
 			if err != nil {
-				continue
+				common.SysError("failed to get channels by tag: " + err.Error())
+				c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取标签渠道失败，请稍后重试"})
+				return
 			}
-			filtered := make([]*model.Channel, 0)
-			for _, ch := range tagChannels {
-				if statusFilter == common.ChannelStatusEnabled && ch.Status != common.ChannelStatusEnabled {
-					continue
-				}
-				if statusFilter == 0 && ch.Status == common.ChannelStatusEnabled {
-					continue
-				}
-				if typeFilter >= 0 && ch.Type != typeFilter {
-					continue
-				}
-				filtered = append(filtered, ch)
-			}
-			channelData = append(channelData, filtered...)
+			channelData = append(channelData, tagChannels...)
 		}
-		total, _ = model.CountAllTags()
 	} else {
-		baseQuery := model.DB.Model(&model.Channel{})
-		if typeFilter >= 0 {
-			baseQuery = baseQuery.Where("type = ?", typeFilter)
-		}
-		if statusFilter == common.ChannelStatusEnabled {
-			baseQuery = baseQuery.Where("status = ?", common.ChannelStatusEnabled)
-		} else if statusFilter == 0 {
-			baseQuery = baseQuery.Where("status != ?", common.ChannelStatusEnabled)
+		if err := buildChannelListQuery(groupFilter, statusFilter, typeFilter).Count(&total).Error; err != nil {
+			common.SysError("failed to count channels: " + err.Error())
+			c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取渠道数量失败，请稍后重试"})
+			return
 		}
 
-		baseQuery.Count(&total)
-
-		err := sortOptions.Apply(baseQuery).Limit(pageInfo.GetPageSize()).Offset(pageInfo.GetStartIdx()).Omit("key").Find(&channelData).Error
+		err := sortOptions.Apply(buildChannelListQuery(groupFilter, statusFilter, typeFilter)).
+			Limit(pageInfo.GetPageSize()).
+			Offset(pageInfo.GetStartIdx()).
+			Omit("key").
+			Find(&channelData).Error
 		if err != nil {
 			common.SysError("failed to get channels: " + err.Error())
 			c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取渠道列表失败，请稍后重试"})
@@ -144,17 +161,16 @@ func GetAllChannels(c *gin.Context) {
 		clearChannelInfo(datum)
 	}
 
-	countQuery := model.DB.Model(&model.Channel{})
-	if statusFilter == common.ChannelStatusEnabled {
-		countQuery = countQuery.Where("status = ?", common.ChannelStatusEnabled)
-	} else if statusFilter == 0 {
-		countQuery = countQuery.Where("status != ?", common.ChannelStatusEnabled)
-	}
+	countQuery := buildChannelListQuery(groupFilter, statusFilter, -1)
 	var results []struct {
 		Type  int64
 		Count int64
 	}
-	_ = countQuery.Select("type, count(*) as count").Group("type").Find(&results).Error
+	if err := countQuery.Select("type, count(*) as count").Group("type").Find(&results).Error; err != nil {
+		common.SysError("failed to count channel types: " + err.Error())
+		c.JSON(http.StatusOK, gin.H{"success": false, "message": "获取渠道类型统计失败，请稍后重试"})
+		return
+	}
 	typeCounts := make(map[int64]int64)
 	for _, r := range results {
 		typeCounts[r.Type] = r.Count
@@ -262,10 +278,18 @@ func SearchChannels(c *gin.Context) {
 		}
 		for _, tag := range tags {
 			if tag != nil && *tag != "" {
-				tagChannel, err := model.GetChannelsByTag(*tag, idSort, false, sortOptions)
-				if err == nil {
-					channelData = append(channelData, tagChannel...)
+				var tagChannels []*model.Channel
+				err := sortOptions.Apply(buildChannelListQuery(group, -1, -1).Where("tag = ?", *tag)).
+					Omit("key").
+					Find(&tagChannels).Error
+				if err != nil {
+					c.JSON(http.StatusOK, gin.H{
+						"success": false,
+						"message": err.Error(),
+					})
+					return
 				}
+				channelData = append(channelData, tagChannels...)
 			}
 		}
 	} else {
@@ -1194,7 +1218,7 @@ func CopyChannel(c *gin.Context) {
 	}
 
 	// insert
-	if err := model.BatchInsertChannels([]model.Channel{clone}); err != nil {
+	if err := clone.Insert(); err != nil {
 		common.SysError("failed to clone channel: " + err.Error())
 		c.JSON(http.StatusOK, gin.H{"success": false, "message": "复制渠道失败，请稍后重试"})
 		return

controller/channel_test_internal_test.go

@@ -69,3 +69,14 @@ func TestBuildTestLogOtherInjectsTieredInfo(t *testing.T) {
 	require.Equal(t, "base", other["matched_tier"])
 	require.NotEmpty(t, other["expr_b64"])
 }
+
+func TestResolveChannelTestUserIDUsesRequestUser(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
+	ctx.Set("id", 2)
+
+	userID, err := resolveChannelTestUserID(ctx)
+
+	require.NoError(t, err)
+	require.Equal(t, 2, userID)
+}

controller/custom_oauth.go

@@ -501,7 +501,7 @@ func GetUserOAuthBindingsByAdmin(c *gin.Context) {
 	}
 
 	myRole := c.GetInt("role")
-	if myRole <= targetUser.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, targetUser.Role) {
 		common.ApiErrorMsg(c, "no permission")
 		return
 	}
@@ -560,7 +560,7 @@ func UnbindCustomOAuthByAdmin(c *gin.Context) {
 	}
 
 	myRole := c.GetInt("role")
-	if myRole <= targetUser.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, targetUser.Role) {
 		common.ApiErrorMsg(c, "no permission")
 		return
 	}

controller/misc.go

@@ -87,6 +87,9 @@ func GetStatus(c *gin.Context) {
 		"chats":                         setting.Chats,
 		"demo_site_enabled":             operation_setting.DemoSiteEnabled,
 		"self_use_mode_enabled":         operation_setting.SelfUseModeEnabled,
+		"register_enabled":              common.RegisterEnabled,
+		"password_login_enabled":        common.PasswordLoginEnabled,
+		"password_register_enabled":     common.PasswordRegisterEnabled,
 		"default_use_auto_group":        setting.DefaultUseAutoGroup,
 
 		"usd_exchange_rate": operation_setting.USDExchangeRate,

controller/model.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"fmt"
 	"net/http"
+	"strings"
 	"time"
 
 	"github.com/QuantumNous/new-api/common"
@@ -109,9 +110,102 @@ func init() {
 	})
 }
 
-func ListModels(c *gin.Context, modelType int) {
-	userOpenAiModels := make([]dto.OpenAIModels, 0)
+func channelOwnerName(channelType int) string {
+	apiType, success := common.ChannelType2APIType(channelType)
+	if !success {
+		return strings.ToLower(constant.GetChannelTypeName(channelType))
+	}
+	adaptor := relay.GetAdaptor(apiType)
+	if adaptor == nil {
+		return strings.ToLower(constant.GetChannelTypeName(channelType))
+	}
+	adaptor.Init(&relaycommon.RelayInfo{ChannelMeta: &relaycommon.ChannelMeta{
+		ChannelType: channelType,
+	}})
+	if name := strings.TrimSpace(adaptor.GetChannelName()); name != "" {
+		return name
+	}
+	return strings.ToLower(constant.GetChannelTypeName(channelType))
+}
 
+func getPreferredModelOwners(modelNames []string, groups []string) map[string]string {
+	channelTypes, err := model.GetPreferredModelOwnerChannelTypes(modelNames, groups)
+	if err != nil {
+		common.SysLog(fmt.Sprintf("GetPreferredModelOwnerChannelTypes error: %v", err))
+		return map[string]string{}
+	}
+
+	ownerByChannelType := make(map[int]string)
+	owners := make(map[string]string, len(channelTypes))
+	for modelName, channelType := range channelTypes {
+		owner, ok := ownerByChannelType[channelType]
+		if !ok {
+			owner = channelOwnerName(channelType)
+			ownerByChannelType[channelType] = owner
+		}
+		if owner != "" {
+			owners[modelName] = owner
+		}
+	}
+	return owners
+}
+
+func buildOpenAIModel(modelName string, ownerByModel map[string]string) dto.OpenAIModels {
+	var oaiModel dto.OpenAIModels
+	if staticModel, ok := openAIModelsMap[modelName]; ok {
+		oaiModel = staticModel
+	} else {
+		oaiModel = dto.OpenAIModels{
+			Id:      modelName,
+			Object:  "model",
+			Created: 1626777600,
+			OwnedBy: "custom",
+		}
+	}
+	if owner, ok := ownerByModel[modelName]; ok && owner != "" {
+		oaiModel.OwnedBy = owner
+	}
+	oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(modelName)
+	return oaiModel
+}
+
+type modelListGroups struct {
+	userGroup   string
+	tokenGroup  string
+	ownerGroups []string
+}
+
+func getModelListGroups(c *gin.Context) (modelListGroups, error) {
+	tokenGroup := common.GetContextKeyString(c, constant.ContextKeyTokenGroup)
+	userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
+	if userGroup == "" && (tokenGroup == "" || tokenGroup == "auto") {
+		var err error
+		userGroup, err = model.GetUserGroup(c.GetInt("id"), false)
+		if err != nil {
+			return modelListGroups{}, err
+		}
+	}
+
+	if tokenGroup == "auto" {
+		return modelListGroups{
+			userGroup:   userGroup,
+			tokenGroup:  tokenGroup,
+			ownerGroups: service.GetUserAutoGroup(userGroup),
+		}, nil
+	}
+
+	group := userGroup
+	if tokenGroup != "" {
+		group = tokenGroup
+	}
+	return modelListGroups{
+		userGroup:   userGroup,
+		tokenGroup:  tokenGroup,
+		ownerGroups: []string{group},
+	}, nil
+}
+
+func ListModels(c *gin.Context, modelType int) {
 	acceptUnsetRatioModel := operation_setting.SelfUseModeEnabled
 	if !acceptUnsetRatioModel {
 		userId := c.GetInt("id")
@@ -123,6 +217,16 @@ func ListModels(c *gin.Context, modelType int) {
 		}
 	}
 
+	userModelNames := make([]string, 0)
+	groups, err := getModelListGroups(c)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "get user group failed",
+		})
+		return
+	}
+	ownerGroups := groups.ownerGroups
 	modelLimitEnable := common.GetContextKeyBool(c, constant.ContextKeyTokenModelLimitEnabled)
 	if modelLimitEnable {
 		s, ok := common.GetContextKey(c, constant.ContextKeyTokenModelLimit)
@@ -138,37 +242,12 @@ func ListModels(c *gin.Context, modelType int) {
 					continue
 				}
 			}
-			if oaiModel, ok := openAIModelsMap[allowModel]; ok {
-				oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(allowModel)
-				userOpenAiModels = append(userOpenAiModels, oaiModel)
-			} else {
-				userOpenAiModels = append(userOpenAiModels, dto.OpenAIModels{
-					Id:                     allowModel,
-					Object:                 "model",
-					Created:                1626777600,
-					OwnedBy:                "custom",
-					SupportedEndpointTypes: model.GetModelSupportEndpointTypes(allowModel),
-				})
-			}
+			userModelNames = append(userModelNames, allowModel)
 		}
 	} else {
-		userId := c.GetInt("id")
-		userGroup, err := model.GetUserGroup(userId, false)
-		if err != nil {
-			c.JSON(http.StatusOK, gin.H{
-				"success": false,
-				"message": "get user group failed",
-			})
-			return
-		}
-		group := userGroup
-		tokenGroup := common.GetContextKeyString(c, constant.ContextKeyTokenGroup)
-		if tokenGroup != "" {
-			group = tokenGroup
-		}
 		var models []string
-		if tokenGroup == "auto" {
-			for _, autoGroup := range service.GetUserAutoGroup(userGroup) {
+		if groups.tokenGroup == "auto" {
+			for _, autoGroup := range ownerGroups {
 				groupModels := model.GetGroupEnabledModels(autoGroup)
 				for _, g := range groupModels {
 					if !common.StringsContains(models, g) {
@@ -177,7 +256,7 @@ func ListModels(c *gin.Context, modelType int) {
 				}
 			}
 		} else {
-			models = model.GetGroupEnabledModels(group)
+			models = model.GetGroupEnabledModels(ownerGroups[0])
 		}
 		for _, modelName := range models {
 			if !acceptUnsetRatioModel {
@@ -185,21 +264,19 @@ func ListModels(c *gin.Context, modelType int) {
 					continue
 				}
 			}
-			if oaiModel, ok := openAIModelsMap[modelName]; ok {
-				oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(modelName)
-				userOpenAiModels = append(userOpenAiModels, oaiModel)
-			} else {
-				userOpenAiModels = append(userOpenAiModels, dto.OpenAIModels{
-					Id:                     modelName,
-					Object:                 "model",
-					Created:                1626777600,
-					OwnedBy:                "custom",
-					SupportedEndpointTypes: model.GetModelSupportEndpointTypes(modelName),
-				})
-			}
+			userModelNames = append(userModelNames, modelName)
 		}
 	}
 
+	ownerByModel := map[string]string{}
+	if len(ownerGroups) > 0 {
+		ownerByModel = getPreferredModelOwners(userModelNames, ownerGroups)
+	}
+	userOpenAiModels := make([]dto.OpenAIModels, 0, len(userModelNames))
+	for _, modelName := range userModelNames {
+		userOpenAiModels = append(userOpenAiModels, buildOpenAIModel(modelName, ownerByModel))
+	}
+
 	switch modelType {
 	case constant.ChannelTypeAnthropic:
 		useranthropicModels := make([]dto.AnthropicModel, len(userOpenAiModels))

controller/model_owned_by_test.go

@@ -0,0 +1,85 @@
+package controller
+
+import (
+	"net/http/httptest"
+	"testing"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/constant"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+func TestChannelOwnerNameUsesAdaptorChannelName(t *testing.T) {
+	tests := []struct {
+		name        string
+		channelType int
+		expected    string
+	}{
+		{
+			name:        "openai",
+			channelType: constant.ChannelTypeOpenAI,
+			expected:    "openai",
+		},
+		{
+			name:        "codex",
+			channelType: constant.ChannelTypeCodex,
+			expected:    "codex",
+		},
+		{
+			name:        "openrouter",
+			channelType: constant.ChannelTypeOpenRouter,
+			expected:    "openrouter",
+		},
+		{
+			name:        "azure fallback",
+			channelType: constant.ChannelTypeAzure,
+			expected:    "azure",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			require.Equal(t, tt.expected, channelOwnerName(tt.channelType))
+		})
+	}
+}
+
+func TestBuildOpenAIModelOverridesOwnedBy(t *testing.T) {
+	modelItem := buildOpenAIModel("gpt-5.4", map[string]string{"gpt-5.4": "openai"})
+	require.Equal(t, "gpt-5.4", modelItem.Id)
+	require.Equal(t, "openai", modelItem.OwnedBy)
+}
+
+func TestBuildOpenAIModelFallsBackToCustomForUnknownModels(t *testing.T) {
+	modelItem := buildOpenAIModel("custom-test-model", nil)
+	require.Equal(t, "custom-test-model", modelItem.Id)
+	require.Equal(t, "custom", modelItem.OwnedBy)
+}
+
+func TestGetModelListGroupsUsesUserGroupWhenTokenGroupIsEmpty(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
+	common.SetContextKey(ctx, constant.ContextKeyUserGroup, "default")
+
+	groups, err := getModelListGroups(ctx)
+	require.NoError(t, err)
+
+	require.Equal(t, "default", groups.userGroup)
+	require.Empty(t, groups.tokenGroup)
+	require.Equal(t, []string{"default"}, groups.ownerGroups)
+}
+
+func TestGetModelListGroupsUsesExplicitTokenGroup(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+	ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
+	common.SetContextKey(ctx, constant.ContextKeyUserGroup, "default")
+	common.SetContextKey(ctx, constant.ContextKeyTokenGroup, "vip")
+
+	groups, err := getModelListGroups(ctx)
+	require.NoError(t, err)
+
+	require.Equal(t, "default", groups.userGroup)
+	require.Equal(t, "vip", groups.tokenGroup)
+	require.Equal(t, []string{"vip"}, groups.ownerGroups)
+}

controller/option.go

@@ -42,15 +42,6 @@ func isPositiveOptionValue(value string) bool {
 	return err == nil && floatValue > 0
 }
 
-func isVisiblePublicKeyOption(key string) bool {
-	switch key {
-	case "WaffoPancakeWebhookPublicKey", "WaffoPancakeWebhookTestKey":
-		return true
-	default:
-		return false
-	}
-}
-
 func collectModelNamesFromOptionValue(raw string, modelNames map[string]struct{}) {
 	if strings.TrimSpace(raw) == "" {
 		return
@@ -95,7 +86,7 @@ func GetOptions(c *gin.Context) {
 			strings.HasSuffix(k, "Key") ||
 			strings.HasSuffix(k, "secret") ||
 			strings.HasSuffix(k, "api_key")
-		if isSensitiveKey && !isVisiblePublicKeyOption(k) {
+		if isSensitiveKey {
 			continue
 		}
 		options = append(options, &model.Option{

controller/passkey.go

@@ -350,6 +350,11 @@ func AdminResetPasskey(c *gin.Context) {
 		common.ApiError(c, err)
 		return
 	}
+	myRole := c.GetInt("role")
+	if !canManageTargetRole(myRole, user.Role) {
+		common.ApiErrorMsg(c, "no permission")
+		return
+	}
 
 	if _, err := model.GetPasskeyByUserID(user.Id); err != nil {
 		if errors.Is(err, model.ErrPasskeyNotFound) {

controller/payment_webhook_availability.go

@@ -77,24 +77,15 @@ func isWaffoPancakeTopUpEnabled() bool {
 	if !isPaymentComplianceConfirmed() {
 		return false
 	}
-	if !setting.WaffoPancakeEnabled {
-		return false
-	}
-
-	return isWaffoPancakeWebhookConfigured() &&
-		strings.TrimSpace(setting.WaffoPancakeMerchantID) != "" &&
+	// Presence-of-credentials = enabled. Webhook public keys ship inside
+	// the SDK; mode (test/prod) is read from each event.
+	return strings.TrimSpace(setting.WaffoPancakeMerchantID) != "" &&
 		strings.TrimSpace(setting.WaffoPancakePrivateKey) != "" &&
-		strings.TrimSpace(setting.WaffoPancakeStoreID) != "" &&
 		strings.TrimSpace(setting.WaffoPancakeProductID) != ""
 }
 
 func isWaffoPancakeWebhookConfigured() bool {
-	currentWebhookKey := strings.TrimSpace(setting.WaffoPancakeWebhookPublicKey)
-	if setting.WaffoPancakeSandbox {
-		currentWebhookKey = strings.TrimSpace(setting.WaffoPancakeWebhookTestKey)
-	}
-
-	return currentWebhookKey != ""
+	return isWaffoPancakeTopUpEnabled()
 }
 
 func isWaffoPancakeWebhookEnabled() bool {

controller/payment_webhook_availability_test.go

@@ -114,47 +114,32 @@ func TestWaffoWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
 
 func TestWaffoPancakeWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
 	confirmPaymentComplianceForTest(t)
-	originalEnabled := setting.WaffoPancakeEnabled
-	originalSandbox := setting.WaffoPancakeSandbox
 	originalMerchantID := setting.WaffoPancakeMerchantID
 	originalPrivateKey := setting.WaffoPancakePrivateKey
-	originalWebhookPublicKey := setting.WaffoPancakeWebhookPublicKey
-	originalWebhookTestKey := setting.WaffoPancakeWebhookTestKey
-	originalStoreID := setting.WaffoPancakeStoreID
 	originalProductID := setting.WaffoPancakeProductID
 	t.Cleanup(func() {
-		setting.WaffoPancakeEnabled = originalEnabled
-		setting.WaffoPancakeSandbox = originalSandbox
 		setting.WaffoPancakeMerchantID = originalMerchantID
 		setting.WaffoPancakePrivateKey = originalPrivateKey
-		setting.WaffoPancakeWebhookPublicKey = originalWebhookPublicKey
-		setting.WaffoPancakeWebhookTestKey = originalWebhookTestKey
-		setting.WaffoPancakeStoreID = originalStoreID
 		setting.WaffoPancakeProductID = originalProductID
 	})
 
-	setting.WaffoPancakeEnabled = true
-	setting.WaffoPancakeSandbox = false
-	setting.WaffoPancakeMerchantID = "merchant"
+	// Presence of all three credentials enables the gateway. Webhook public
+	// keys are bundled in the SDK and there is no separate Enabled toggle —
+	// clear any of the three fields to disable.
+	setting.WaffoPancakeMerchantID = ""
 	setting.WaffoPancakePrivateKey = "private"
-	setting.WaffoPancakeStoreID = "store"
 	setting.WaffoPancakeProductID = "product"
-	setting.WaffoPancakeWebhookPublicKey = ""
 	require.False(t, isWaffoPancakeWebhookEnabled())
 
-	setting.WaffoPancakeWebhookPublicKey = "public"
+	setting.WaffoPancakeMerchantID = "merchant"
 	require.True(t, isWaffoPancakeWebhookEnabled())
 
-	setting.WaffoPancakeEnabled = false
+	setting.WaffoPancakeProductID = ""
 	require.False(t, isWaffoPancakeWebhookEnabled())
 
-	setting.WaffoPancakeEnabled = true
-	setting.WaffoPancakeSandbox = true
-	setting.WaffoPancakeWebhookTestKey = ""
+	setting.WaffoPancakeProductID = "product"
+	setting.WaffoPancakePrivateKey = ""
 	require.False(t, isWaffoPancakeWebhookEnabled())
-
-	setting.WaffoPancakeWebhookTestKey = "test_public"
-	require.True(t, isWaffoPancakeWebhookEnabled())
 }
 
 func TestEpayWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {

controller/perf_metrics.go

@@ -8,6 +8,7 @@ import (
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
 
 	"github.com/gin-gonic/gin"
+	"github.com/samber/lo"
 )
 
 func GetPerfMetricsSummary(c *gin.Context) {
@@ -18,7 +19,8 @@ func GetPerfMetricsSummary(c *gin.Context) {
 		}
 	}
 
-	result, err := perfmetrics.QuerySummaryAll(hours)
+	activeGroups := append(lo.Keys(ratio_setting.GetGroupRatioCopy()), "auto")
+	result, err := perfmetrics.QuerySummaryAll(hours, activeGroups)
 	if err != nil {
 		c.JSON(http.StatusInternalServerError, gin.H{
 			"success": false,
@@ -72,12 +74,9 @@ func GetPerfMetrics(c *gin.Context) {
 }
 
 func filterActiveGroups(groups []perfmetrics.GroupResult) []perfmetrics.GroupResult {
-	activeGroups := ratio_setting.GetGroupRatioCopy()
-	filtered := make([]perfmetrics.GroupResult, 0, len(groups))
-	for _, g := range groups {
-		if _, ok := activeGroups[g.Group]; ok || g.Group == "auto" {
-			filtered = append(filtered, g)
-		}
-	}
-	return filtered
+	activeRatios := ratio_setting.GetGroupRatioCopy()
+	return lo.Filter(groups, func(g perfmetrics.GroupResult, _ int) bool {
+		_, ok := activeRatios[g.Group]
+		return ok || g.Group == "auto"
+	})
 }

controller/rankings.go

@@ -3,51 +3,11 @@ package controller
 import (
 	"net/http"
 
-	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/gin-gonic/gin"
 )
 
-func isRankingsEnabled() bool {
-	common.OptionMapRWMutex.RLock()
-	raw := common.OptionMap["HeaderNavModules"]
-	common.OptionMapRWMutex.RUnlock()
-
-	if raw == "" {
-		return true
-	}
-
-	var parsed map[string]interface{}
-	if err := common.Unmarshal([]byte(raw), &parsed); err != nil {
-		return true
-	}
-	rankings, ok := parsed["rankings"]
-	if !ok {
-		return true
-	}
-	switch v := rankings.(type) {
-	case bool:
-		return v
-	case map[string]interface{}:
-		if enabled, ok := v["enabled"]; ok {
-			if b, ok := enabled.(bool); ok {
-				return b
-			}
-		}
-		return true
-	}
-	return true
-}
-
 func GetRankings(c *gin.Context) {
-	if !isRankingsEnabled() {
-		c.JSON(http.StatusForbidden, gin.H{
-			"success": false,
-			"message": "rankings is disabled",
-		})
-		return
-	}
-
 	result, err := service.GetRankingsSnapshot(c.DefaultQuery("period", "week"))
 	if err != nil {
 		c.JSON(http.StatusBadRequest, gin.H{

controller/relay.go

@@ -88,7 +88,7 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
 
 	defer func() {
 		if newAPIError != nil {
-			logger.LogError(c, fmt.Sprintf("relay error: %s", newAPIError.Error()))
+			logger.LogError(c, fmt.Sprintf("relay error: %s", common.LocalLogPreview(newAPIError.Error())))
 			newAPIError.SetMessage(common.MessageWithRequestId(newAPIError.Error(), requestId))
 			switch relayFormat {
 			case types.RelayFormatOpenAIRealtime:
@@ -354,7 +354,7 @@ func shouldRetry(c *gin.Context, openaiErr *types.NewAPIError, retryTimes int) b
 }
 
 func processChannelError(c *gin.Context, channelError types.ChannelError, err *types.NewAPIError) {
-	logger.LogError(c, fmt.Sprintf("channel error (channel #%d, status code: %d): %s", channelError.ChannelId, err.StatusCode, err.Error()))
+	logger.LogError(c, fmt.Sprintf("channel error (channel #%d, status code: %d): %s", channelError.ChannelId, err.StatusCode, common.LocalLogPreview(err.Error())))
 	// 不要使用context获取渠道信息，异步处理时可能会出现渠道信息不一致的情况
 	// do not use context to get channel info, there may be inconsistent channel info when processing asynchronously
 	if service.ShouldDisableChannel(err) && channelError.AutoBan {

controller/subscription.go

@@ -22,6 +22,10 @@ type BillingPreferenceRequest struct {
 	BillingPreference string `json:"billing_preference"`
 }
 
+type SubscriptionBalancePayRequest struct {
+	PlanId int `json:"plan_id"`
+}
+
 // ---- User APIs ----
 
 func GetSubscriptionPlans(c *gin.Context) {
@@ -37,6 +41,7 @@ func GetSubscriptionPlans(c *gin.Context) {
 	}
 	result := make([]SubscriptionPlanDTO, 0, len(plans))
 	for _, p := range plans {
+		p.NormalizeDefaults()
 		result = append(result, SubscriptionPlanDTO{
 			Plan: p,
 		})
@@ -92,6 +97,25 @@ func UpdateSubscriptionPreference(c *gin.Context) {
 	common.ApiSuccess(c, gin.H{"billing_preference": pref})
 }
 
+func SubscriptionRequestBalancePay(c *gin.Context) {
+	if !requirePaymentCompliance(c) {
+		return
+	}
+
+	userId := c.GetInt("id")
+	var req SubscriptionBalancePayRequest
+	if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
+		common.ApiErrorMsg(c, "参数错误")
+		return
+	}
+
+	if err := model.PurchaseSubscriptionWithBalance(userId, req.PlanId); err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	common.ApiSuccess(c, nil)
+}
+
 // ---- Admin APIs ----
 
 func AdminListSubscriptionPlans(c *gin.Context) {
@@ -102,6 +126,7 @@ func AdminListSubscriptionPlans(c *gin.Context) {
 	}
 	result := make([]SubscriptionPlanDTO, 0, len(plans))
 	for _, p := range plans {
+		p.NormalizeDefaults()
 		result = append(result, SubscriptionPlanDTO{
 			Plan: p,
 		})
@@ -140,6 +165,9 @@ func AdminCreateSubscriptionPlan(c *gin.Context) {
 		req.Plan.Currency = "USD"
 	}
 	req.Plan.Currency = "USD"
+	if req.Plan.AllowBalancePay == nil {
+		req.Plan.AllowBalancePay = common.GetPointer(true)
+	}
 	if req.Plan.DurationUnit == "" {
 		req.Plan.DurationUnit = model.SubscriptionDurationMonth
 	}
@@ -248,6 +276,7 @@ func AdminUpdateSubscriptionPlan(c *gin.Context) {
 			"sort_order":                 req.Plan.SortOrder,
 			"stripe_price_id":            req.Plan.StripePriceId,
 			"creem_product_id":           req.Plan.CreemProductId,
+			"waffo_pancake_product_id":   req.Plan.WaffoPancakeProductId,
 			"max_purchase_per_user":      req.Plan.MaxPurchasePerUser,
 			"total_amount":               req.Plan.TotalAmount,
 			"upgrade_group":              req.Plan.UpgradeGroup,
@@ -255,6 +284,9 @@ func AdminUpdateSubscriptionPlan(c *gin.Context) {
 			"quota_reset_custom_seconds": req.Plan.QuotaResetCustomSeconds,
 			"updated_at":                 common.GetTimestamp(),
 		}
+		if req.Plan.AllowBalancePay != nil {
+			updateMap["allow_balance_pay"] = *req.Plan.AllowBalancePay
+		}
 		if err := tx.Model(&model.SubscriptionPlan{}).Where("id = ?", id).Updates(updateMap).Error; err != nil {
 			return err
 		}

controller/subscription_payment_waffo_pancake.go

@@ -0,0 +1,130 @@
+package controller
+
+import (
+	"fmt"
+	"net/http"
+	"strings"
+	"time"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/logger"
+	"github.com/QuantumNous/new-api/model"
+	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/setting"
+	"github.com/gin-gonic/gin"
+	"github.com/shopspring/decimal"
+	"github.com/thanhpk/randstr"
+)
+
+type SubscriptionWaffoPancakePayRequest struct {
+	PlanId int `json:"plan_id"`
+}
+
+func SubscriptionRequestWaffoPancakePay(c *gin.Context) {
+	if !requirePaymentCompliance(c) {
+		return
+	}
+
+	var req SubscriptionWaffoPancakePayRequest
+	if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
+		common.ApiErrorMsg(c, "参数错误")
+		return
+	}
+
+	plan, err := model.GetSubscriptionPlanById(req.PlanId)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	if !plan.Enabled {
+		common.ApiErrorMsg(c, "套餐未启用")
+		return
+	}
+	if strings.TrimSpace(plan.WaffoPancakeProductId) == "" {
+		common.ApiErrorMsg(c, "该套餐未配置 WaffoPancakeProductId")
+		return
+	}
+	// Plan targets its own Pancake product, so we only require credentials
+	// here — not the gateway-level WaffoPancakeProductID.
+	if strings.TrimSpace(setting.WaffoPancakeMerchantID) == "" ||
+		strings.TrimSpace(setting.WaffoPancakePrivateKey) == "" {
+		common.ApiErrorMsg(c, "Waffo Pancake 未配置或密钥无效")
+		return
+	}
+
+	userId := c.GetInt("id")
+	user, err := model.GetUserById(userId, false)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	if user == nil {
+		common.ApiErrorMsg(c, "用户不存在")
+		return
+	}
+
+	if plan.MaxPurchasePerUser > 0 {
+		count, err := model.CountUserSubscriptionsByPlan(userId, plan.Id)
+		if err != nil {
+			common.ApiError(c, err)
+			return
+		}
+		if count >= int64(plan.MaxPurchasePerUser) {
+			common.ApiErrorMsg(c, "已达到该套餐购买上限")
+			return
+		}
+	}
+
+	// WAFFO_PANCAKE_SUB- prefix (vs. wallet's WAFFO_PANCAKE-) drives webhook
+	// dispatch in WaffoPancakeWebhook.
+	tradeNo := fmt.Sprintf("WAFFO_PANCAKE_SUB-%d-%d-%s", userId, time.Now().UnixMilli(), randstr.String(6))
+
+	order := &model.SubscriptionOrder{
+		UserId:          userId,
+		PlanId:          plan.Id,
+		Money:           plan.PriceAmount,
+		TradeNo:         tradeNo,
+		PaymentMethod:   model.PaymentMethodWaffoPancake,
+		PaymentProvider: model.PaymentProviderWaffoPancake,
+		CreateTime:      time.Now().Unix(),
+		Status:          common.TopUpStatusPending,
+	}
+	if err := order.Insert(); err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅订单创建失败 user_id=%d plan_id=%d trade_no=%s error=%q", userId, plan.Id, tradeNo, err.Error()))
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "创建订单失败"})
+		return
+	}
+
+	expiresInSeconds := 45 * 60
+	session, err := service.CreateWaffoPancakeCheckoutSession(c.Request.Context(), &service.WaffoPancakeCreateSessionParams{
+		ProductID:     plan.WaffoPancakeProductId,
+		BuyerIdentity: service.WaffoPancakeBuyerIdentityFromUserID(user.Id),
+		PriceSnapshot: &service.WaffoPancakePriceSnapshot{
+			Amount:      decimal.NewFromFloat(plan.PriceAmount).StringFixed(2),
+			TaxCategory: "saas",
+		},
+		BuyerEmail:              getWaffoPancakeBuyerEmail(user),
+		ExpiresInSeconds:        &expiresInSeconds,
+		OrderMerchantExternalID: tradeNo,
+	})
+	if err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅结账会话创建失败 user_id=%d plan_id=%d trade_no=%s error=%q", userId, plan.Id, tradeNo, err.Error()))
+		order.Status = common.TopUpStatusFailed
+		_ = order.Update()
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉起支付失败"})
+		return
+	}
+	logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅订单创建成功 user_id=%d plan_id=%d trade_no=%s session_id=%s money=%.2f", userId, plan.Id, tradeNo, session.SessionID, plan.PriceAmount))
+
+	c.JSON(http.StatusOK, gin.H{
+		"message": "success",
+		"data": gin.H{
+			"checkout_url":     session.CheckoutURL,
+			"session_id":       session.SessionID,
+			"expires_at":       session.ExpiresAt,
+			"order_id":         tradeNo,
+			"token":            session.Token,
+			"token_expires_at": session.TokenExpiresAt,
+		},
+	})
+}

controller/task_video.go

@@ -96,13 +96,13 @@ func updateVideoSingleTask(ctx context.Context, adaptor channel.TaskAdaptor, cha
 		return fmt.Errorf("readAll failed for task %s: %w", taskId, err)
 	}
 
-	logger.LogDebug(ctx, fmt.Sprintf("UpdateVideoSingleTask response: %s", string(responseBody)))
+	logger.LogDebug(ctx, "UpdateVideoSingleTask response: %s", responseBody)
 
 	taskResult := &relaycommon.TaskInfo{}
 	// try parse as New API response format
 	var responseItems dto.TaskResponse[model.Task]
 	if err = common.Unmarshal(responseBody, &responseItems); err == nil && responseItems.IsSuccess() {
-		logger.LogDebug(ctx, fmt.Sprintf("UpdateVideoSingleTask parsed as new api response format: %+v", responseItems))
+		logger.LogDebug(ctx, "UpdateVideoSingleTask parsed as new api response format: %+v", responseItems)
 		t := responseItems.Data
 		taskResult.TaskID = t.TaskID
 		taskResult.Status = string(t.Status)
@@ -116,7 +116,7 @@ func updateVideoSingleTask(ctx context.Context, adaptor channel.TaskAdaptor, cha
 		task.Data = redactVideoResponseBody(responseBody)
 	}
 
-	logger.LogDebug(ctx, fmt.Sprintf("UpdateVideoSingleTask taskResult: %+v", taskResult))
+	logger.LogDebug(ctx, "UpdateVideoSingleTask taskResult: %+v", taskResult)
 
 	now := time.Now().Unix()
 	if taskResult.Status == "" {

controller/topup.go

@@ -52,6 +52,27 @@ func GetTopUpInfo(c *gin.Context) {
 		}
 	}
 
+	// Waffo Pancake displayed above the legacy Waffo gateway.
+	enableWaffoPancake := isWaffoPancakeTopUpEnabled()
+	if enableWaffoPancake {
+		hasWaffoPancake := false
+		for _, method := range payMethods {
+			if method["type"] == model.PaymentMethodWaffoPancake {
+				hasWaffoPancake = true
+				break
+			}
+		}
+
+		if !hasWaffoPancake {
+			payMethods = append(payMethods, map[string]string{
+				"name":      "Waffo Pancake",
+				"type":      model.PaymentMethodWaffoPancake,
+				"color":     "rgba(var(--semi-orange-5), 1)",
+				"min_topup": strconv.Itoa(setting.WaffoPancakeMinTopUp),
+			})
+		}
+	}
+
 	// 如果启用了 Waffo 支付，添加到支付方法列表
 	enableWaffo := isWaffoTopUpEnabled()
 	if enableWaffo {
@@ -74,26 +95,6 @@ func GetTopUpInfo(c *gin.Context) {
 		}
 	}
 
-	enableWaffoPancake := isWaffoPancakeTopUpEnabled()
-	if enableWaffoPancake {
-		hasWaffoPancake := false
-		for _, method := range payMethods {
-			if method["type"] == model.PaymentMethodWaffoPancake {
-				hasWaffoPancake = true
-				break
-			}
-		}
-
-		if !hasWaffoPancake {
-			payMethods = append(payMethods, map[string]string{
-				"name":      "Waffo Pancake",
-				"type":      model.PaymentMethodWaffoPancake,
-				"color":     "rgba(var(--semi-orange-5), 1)",
-				"min_topup": strconv.Itoa(setting.WaffoPancakeMinTopUp),
-			})
-		}
-	}
-
 	data := gin.H{
 		"enable_online_topup":              isEpayTopUpEnabled(),
 		"enable_stripe_topup":              isStripeTopUpEnabled(),

controller/topup_waffo_pancake.go

@@ -96,33 +96,257 @@ func getWaffoPancakeBuyerEmail(user *model.User) string {
 	if user != nil && strings.TrimSpace(user.Email) != "" {
 		return user.Email
 	}
-	if user != nil {
-		return fmt.Sprintf("%d@new-api.local", user.Id)
-	}
 	return ""
 }
 
-func getWaffoPancakeReturnURL() string {
-	if strings.TrimSpace(setting.WaffoPancakeReturnURL) != "" {
-		return setting.WaffoPancakeReturnURL
+// The admin config endpoints below accept typed-but-not-yet-saved creds in
+// the body and fall back to persisted creds when the body is blank (see
+// resolveWaffoPancakeAdminCreds). Only SaveWaffoPancake writes to OptionMap.
+
+type waffoPancakeCredsRequest struct {
+	MerchantID string `json:"merchant_id"`
+	PrivateKey string `json:"private_key"`
+}
+
+type saveWaffoPancakeRequest struct {
+	MerchantID string `json:"merchant_id"`
+	PrivateKey string `json:"private_key"`
+	ReturnURL  string `json:"return_url"`
+	StoreID    string `json:"store_id"`
+	ProductID  string `json:"product_id"`
+}
+
+type createWaffoPancakePairRequest struct {
+	MerchantID string `json:"merchant_id"`
+	PrivateKey string `json:"private_key"`
+	ReturnURL  string `json:"return_url"`
+}
+
+// SaveWaffoPancake atomically persists all five operator-controlled fields.
+// Catalog / pair endpoints are transient — only this one writes the OptionMap.
+func SaveWaffoPancake(c *gin.Context) {
+	var req saveWaffoPancakeRequest
+	if err := c.ShouldBindJSON(&req); err != nil {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
+		return
 	}
-	return paymentReturnPath("/console/topup?show_history=true")
+	if err := service.SaveWaffoPancakeConfig(
+		c.Request.Context(),
+		req.MerchantID,
+		req.PrivateKey,
+		req.ReturnURL,
+		req.StoreID,
+		req.ProductID,
+	); err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake 保存配置失败 store_id=%q product_id=%q error=%q",
+			req.StoreID, req.ProductID, err.Error(),
+		))
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "保存配置失败"})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"message": "success",
+		"data": gin.H{
+			"product_id": setting.WaffoPancakeProductID,
+			"store_id":   setting.WaffoPancakeStoreID,
+		},
+	})
+}
+
+// resolveWaffoPancakeAdminCreds prefers body creds (typed-but-not-yet-saved
+// values, for verification) and falls back to persisted creds when the body
+// is blank (so returning admins don't have to re-paste the private key,
+// which is stripped from GET /api/option/).
+func resolveWaffoPancakeAdminCreds(bodyMerchantID, bodyPrivateKey string) (string, string) {
+	m := strings.TrimSpace(bodyMerchantID)
+	k := strings.TrimSpace(bodyPrivateKey)
+	if m == "" && k == "" {
+		return setting.WaffoPancakeMerchantID, setting.WaffoPancakePrivateKey
+	}
+	return m, k
+}
+
+// CreateWaffoPancakePair mints a Store + OnetimeProduct pair in one round-
+// trip. Surfaces an orphan-store flag when the product half fails so the
+// frontend can preselect / retry without losing context.
+func CreateWaffoPancakePair(c *gin.Context) {
+	var req createWaffoPancakePairRequest
+	if c.Request.ContentLength > 0 {
+		if err := c.ShouldBindJSON(&req); err != nil {
+			c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
+			return
+		}
+	}
+	merchantID, privateKey := resolveWaffoPancakeAdminCreds(req.MerchantID, req.PrivateKey)
+	if merchantID == "" || privateKey == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 凭证未配置"})
+		return
+	}
+	result, err := service.CreateWaffoPancakePrimaryPair(
+		c.Request.Context(), merchantID, privateKey, req.ReturnURL,
+	)
+	if err != nil {
+		orphan := result != nil && result.OrphanStore
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake 创建店铺与产品失败 orphan_store=%t store_id=%q error=%q",
+			orphan, func() string {
+				if result == nil {
+					return ""
+				}
+				return result.StoreID
+			}(), err.Error(),
+		))
+		data := gin.H{"error": err.Error()}
+		if orphan {
+			data["store_id"] = result.StoreID
+			data["store_name"] = result.StoreName
+			data["orphan_store"] = true
+		}
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": data})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"message": "success",
+		"data": gin.H{
+			"store_id":     result.StoreID,
+			"store_name":   result.StoreName,
+			"product_id":   result.ProductID,
+			"product_name": result.ProductName,
+		},
+	})
+}
+
+// ListWaffoPancakeCatalog returns the merchant's Stores + OnetimeProducts.
+// Doubles as a credential probe (a successful 200 proves the resolved creds
+// authenticate). See resolveWaffoPancakeAdminCreds for credential resolution.
+func ListWaffoPancakeCatalog(c *gin.Context) {
+	var req waffoPancakeCredsRequest
+	// An empty body means "use persisted creds"; only fail on malformed JSON.
+	if c.Request.ContentLength > 0 {
+		if err := c.ShouldBindJSON(&req); err != nil {
+			c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
+			return
+		}
+	}
+	merchantID, privateKey := resolveWaffoPancakeAdminCreds(req.MerchantID, req.PrivateKey)
+	if merchantID == "" || privateKey == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 凭证未配置"})
+		return
+	}
+	catalog, err := service.ListWaffoPancakeCatalog(c.Request.Context(), merchantID, privateKey)
+	if err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake 拉取店铺与产品目录失败 error=%q", err.Error(),
+		))
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉取目录失败"})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{"message": "success", "data": catalog})
+}
+
+type createWaffoPancakeSubscriptionProductRequest struct {
+	Name   string `json:"name"`
+	Amount string `json:"amount"`
+}
+
+// CreateWaffoPancakeSubscriptionProduct mints an OnetimeProduct (not
+// SubscriptionProduct — see service.CreateWaffoPancakeProductForPlan)
+// sized to a plan's `name` + `amount`, using persisted Pancake credentials
+// + StoreID. Reads from the form, not the plan row, so newly-typed unsaved
+// plans can mint a product too.
+func CreateWaffoPancakeSubscriptionProduct(c *gin.Context) {
+	var req createWaffoPancakeSubscriptionProductRequest
+	if c.Request.ContentLength > 0 {
+		if err := c.ShouldBindJSON(&req); err != nil {
+			c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
+			return
+		}
+	}
+	if strings.TrimSpace(req.Name) == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "套餐名称不能为空"})
+		return
+	}
+	if strings.TrimSpace(req.Amount) == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "套餐价格不能为空"})
+		return
+	}
+	merchantID, privateKey := resolveWaffoPancakeAdminCreds("", "")
+	storeID := strings.TrimSpace(setting.WaffoPancakeStoreID)
+	if merchantID == "" || privateKey == "" || storeID == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 未完成配置，请先在支付设置中完成网关绑定"})
+		return
+	}
+	productID, err := service.CreateWaffoPancakeProductForPlan(
+		c.Request.Context(),
+		merchantID,
+		privateKey,
+		storeID,
+		req.Name,
+		req.Amount,
+		setting.WaffoPancakeReturnURL,
+	)
+	if err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake 创建套餐产品失败 store_id=%q name=%q amount=%q error=%q",
+			storeID, req.Name, req.Amount, err.Error(),
+		))
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "创建套餐产品失败"})
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"message": "success",
+		"data": gin.H{
+			"product_id":   productID,
+			"product_name": req.Name,
+			"store_id":     storeID,
+		},
+	})
+}
+
+// ListWaffoPancakeSubscriptionProductOptions returns the OnetimeProducts
+// in the saved Pancake store, for the subscription-plan dropdown. The name
+// reflects new-api's plan concept; under the hood it's still OnetimeProducts.
+func ListWaffoPancakeSubscriptionProductOptions(c *gin.Context) {
+	merchantID, privateKey := resolveWaffoPancakeAdminCreds("", "")
+	storeID := strings.TrimSpace(setting.WaffoPancakeStoreID)
+	if merchantID == "" || privateKey == "" || storeID == "" {
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 未完成配置，请先在支付设置中完成网关绑定"})
+		return
+	}
+	catalog, err := service.ListWaffoPancakeCatalog(c.Request.Context(), merchantID, privateKey)
+	if err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake 拉取订阅产品列表失败 store_id=%q error=%q", storeID, err.Error(),
+		))
+		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉取产品列表失败"})
+		return
+	}
+	products := []service.WaffoPancakeCatalogProduct{}
+	for _, store := range catalog.Stores {
+		if store.ID == storeID {
+			products = store.OnetimeProducts
+			break
+		}
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"message": "success",
+		"data": gin.H{
+			"store_id": storeID,
+			"products": products,
+		},
+	})
+}
+
+func getWaffoPancakeBuyerIdentity(user *model.User) string {
+	if user == nil {
+		return ""
+	}
+	return service.WaffoPancakeBuyerIdentityFromUserID(user.Id)
 }
 
 func RequestWaffoPancakePay(c *gin.Context) {
-	if !setting.WaffoPancakeEnabled {
-		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 支付未启用"})
-		return
-	}
-	currentWebhookKey := setting.WaffoPancakeWebhookPublicKey
-	if setting.WaffoPancakeSandbox {
-		currentWebhookKey = setting.WaffoPancakeWebhookTestKey
-	}
-	if strings.TrimSpace(setting.WaffoPancakeMerchantID) == "" ||
-		strings.TrimSpace(setting.WaffoPancakePrivateKey) == "" ||
-		strings.TrimSpace(currentWebhookKey) == "" ||
-		strings.TrimSpace(setting.WaffoPancakeStoreID) == "" ||
-		strings.TrimSpace(setting.WaffoPancakeProductID) == "" {
+	if !isWaffoPancakeTopUpEnabled() {
 		c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 配置不完整"})
 		return
 	}
@@ -175,18 +399,15 @@ func RequestWaffoPancakePay(c *gin.Context) {
 
 	expiresInSeconds := 45 * 60
 	session, err := service.CreateWaffoPancakeCheckoutSession(c.Request.Context(), &service.WaffoPancakeCreateSessionParams{
-		StoreID:     setting.WaffoPancakeStoreID,
-		ProductID:   setting.WaffoPancakeProductID,
-		ProductType: "onetime",
-		Currency:    strings.ToUpper(strings.TrimSpace(setting.WaffoPancakeCurrency)),
+		ProductID:     setting.WaffoPancakeProductID,
+		BuyerIdentity: getWaffoPancakeBuyerIdentity(user),
 		PriceSnapshot: &service.WaffoPancakePriceSnapshot{
 			Amount:      formatWaffoPancakeAmount(payMoney),
-			TaxIncluded: false,
 			TaxCategory: "saas",
 		},
-		BuyerEmail:       getWaffoPancakeBuyerEmail(user),
-		SuccessURL:       getWaffoPancakeReturnURL(),
-		ExpiresInSeconds: &expiresInSeconds,
+		BuyerEmail:              getWaffoPancakeBuyerEmail(user),
+		ExpiresInSeconds:        &expiresInSeconds,
+		OrderMerchantExternalID: tradeNo,
 	})
 	if err != nil {
 		logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 创建结账会话失败 user_id=%d trade_no=%s error=%q", id, tradeNo, err.Error()))
@@ -200,10 +421,12 @@ func RequestWaffoPancakePay(c *gin.Context) {
 	c.JSON(http.StatusOK, gin.H{
 		"message": "success",
 		"data": gin.H{
-			"checkout_url": session.CheckoutURL,
-			"session_id":   session.SessionID,
-			"expires_at":   session.ExpiresAt,
-			"order_id":     tradeNo,
+			"checkout_url":     session.CheckoutURL,
+			"session_id":       session.SessionID,
+			"expires_at":       session.ExpiresAt,
+			"order_id":         tradeNo,
+			"token":            session.Token,
+			"token_expires_at": session.TokenExpiresAt,
 		},
 	})
 }
@@ -215,6 +438,19 @@ func WaffoPancakeWebhook(c *gin.Context) {
 		return
 	}
 
+	// :env splits test vs prod traffic at the routing layer — operator
+	// registers each URL in the matching webhook slot in Pancake's dashboard.
+	// We then enforce event.mode == expectedEnv to catch mis-registrations.
+	expectedEnv := strings.TrimSpace(c.Param("env"))
+	if expectedEnv != "test" && expectedEnv != "prod" {
+		logger.LogWarn(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake webhook 路径环境段无效 env=%q path=%q client_ip=%s",
+			expectedEnv, c.Request.RequestURI, c.ClientIP(),
+		))
+		c.String(http.StatusNotFound, "unknown env")
+		return
+	}
+
 	bodyBytes, err := io.ReadAll(c.Request.Body)
 	if err != nil {
 		logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 读取请求体失败 path=%q client_ip=%s error=%q", c.Request.RequestURI, c.ClientIP(), err.Error()))
@@ -232,15 +468,57 @@ func WaffoPancakeWebhook(c *gin.Context) {
 		return
 	}
 
+	if !strings.EqualFold(strings.TrimSpace(event.Mode), expectedEnv) {
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake webhook 环境不匹配 expected=%q actual_mode=%q event_id=%s order_id=%s client_ip=%s",
+			expectedEnv, event.Mode, event.ID, event.Data.OrderID, c.ClientIP(),
+		))
+		c.String(http.StatusOK, "OK")
+		return
+	}
+
 	logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 验签成功 event_type=%s event_id=%s order_id=%s client_ip=%s", event.NormalizedEventType(), event.ID, event.Data.OrderID, c.ClientIP()))
 	if event.NormalizedEventType() != "order.completed" {
 		c.String(http.StatusOK, "OK")
 		return
 	}
 
+	// Dispatch by trade_no prefix. OrderMerchantExternalID = our trade_no;
+	// OrderID is Pancake's internal ORD_* (logs only).
+	rawTradeNo := strings.TrimSpace(event.Data.OrderMerchantExternalID)
+	isSubscription := strings.HasPrefix(rawTradeNo, "WAFFO_PANCAKE_SUB-")
+
+	if isSubscription {
+		tradeNo, err := service.ResolveWaffoPancakeSubscriptionTradeNo(event)
+		if err != nil {
+			logger.LogError(c.Request.Context(), fmt.Sprintf(
+				"Waffo Pancake webhook 订阅订单解析失败 event_id=%s order_id=%s buyer_identity=%q client_ip=%s error=%q",
+				event.ID, event.Data.OrderID, event.Data.MerchantProvidedBuyerIdentity, c.ClientIP(), err.Error(),
+			))
+			c.String(http.StatusOK, "OK")
+			return
+		}
+		LockOrder(tradeNo)
+		defer UnlockOrder(tradeNo)
+		if err := model.CompleteSubscriptionOrder(tradeNo, string(bodyBytes), model.PaymentProviderWaffoPancake, ""); err != nil {
+			logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅完成失败 trade_no=%s event_id=%s order_id=%s client_ip=%s error=%q", tradeNo, event.ID, event.Data.OrderID, c.ClientIP(), err.Error()))
+			c.String(http.StatusInternalServerError, "retry")
+			return
+		}
+		logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅完成 trade_no=%s event_id=%s order_id=%s client_ip=%s", tradeNo, event.ID, event.Data.OrderID, c.ClientIP()))
+		c.String(http.StatusOK, "OK")
+		return
+	}
+
 	tradeNo, err := service.ResolveWaffoPancakeTradeNo(event)
 	if err != nil {
-		logger.LogWarn(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 订单号映射失败 event_id=%s order_id=%s error=%q", event.ID, event.Data.OrderID, err.Error()))
+		// LogError (not LogWarn): covers order-not-found and buyer-identity
+		// mismatch — both warrant human attention. 200 OK so Waffo doesn't
+		// retry a permanently-unresolvable webhook.
+		logger.LogError(c.Request.Context(), fmt.Sprintf(
+			"Waffo Pancake webhook 订单解析失败 event_id=%s order_id=%s buyer_identity=%q client_ip=%s error=%q",
+			event.ID, event.Data.OrderID, event.Data.MerchantProvidedBuyerIdentity, c.ClientIP(), err.Error(),
+		))
 		c.String(http.StatusOK, "OK")
 		return
 	}

controller/twofa.go

@@ -520,7 +520,7 @@ func AdminDisable2FA(c *gin.Context) {
 	}
 
 	myRole := c.GetInt("role")
-	if myRole <= targetUser.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, targetUser.Role) {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
 			"message": "无权操作同级或更高级用户的2FA设置",

controller/user.go

@@ -251,8 +251,20 @@ func GetAllUsers(c *gin.Context) {
 func SearchUsers(c *gin.Context) {
 	keyword := c.Query("keyword")
 	group := c.Query("group")
+	var role *int
+	if roleStr := c.Query("role"); roleStr != "" {
+		if parsed, err := strconv.Atoi(roleStr); err == nil {
+			role = &parsed
+		}
+	}
+	var status *int
+	if statusStr := c.Query("status"); statusStr != "" {
+		if parsed, err := strconv.Atoi(statusStr); err == nil {
+			status = &parsed
+		}
+	}
 	pageInfo := common.GetPageQuery(c)
-	users, total, err := model.SearchUsers(keyword, group, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
+	users, total, err := model.SearchUsers(keyword, group, role, status, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
 	if err != nil {
 		common.ApiError(c, err)
 		return
@@ -264,6 +276,10 @@ func SearchUsers(c *gin.Context) {
 	return
 }
 
+func canManageTargetRole(myRole int, targetRole int) bool {
+	return myRole == common.RoleRootUser || myRole > targetRole
+}
+
 func GetUser(c *gin.Context) {
 	id, err := strconv.Atoi(c.Param("id"))
 	if err != nil {
@@ -276,7 +292,7 @@ func GetUser(c *gin.Context) {
 		return
 	}
 	myRole := c.GetInt("role")
-	if myRole <= user.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, user.Role) {
 		common.ApiErrorI18n(c, i18n.MsgUserNoPermissionSameLevel)
 		return
 	}
@@ -567,11 +583,11 @@ func UpdateUser(c *gin.Context) {
 		return
 	}
 	myRole := c.GetInt("role")
-	if myRole <= originUser.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, originUser.Role) {
 		common.ApiErrorI18n(c, i18n.MsgUserNoPermissionHigherLevel)
 		return
 	}
-	if myRole <= updatedUser.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, updatedUser.Role) {
 		common.ApiErrorI18n(c, i18n.MsgUserCannotCreateHigherLevel)
 		return
 	}
@@ -610,7 +626,7 @@ func AdminClearUserBinding(c *gin.Context) {
 	}
 
 	myRole := c.GetInt("role")
-	if myRole <= user.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, user.Role) {
 		common.ApiErrorI18n(c, i18n.MsgUserNoPermissionSameLevel)
 		return
 	}
@@ -778,12 +794,14 @@ func DeleteUser(c *gin.Context) {
 	}
 	err = model.HardDeleteUserById(id)
 	if err != nil {
-		c.JSON(http.StatusOK, gin.H{
-			"success": true,
-			"message": "",
-		})
+		common.ApiError(c, err)
 		return
 	}
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "",
+	})
+	return
 }
 
 func DeleteSelf(c *gin.Context) {
@@ -872,7 +890,7 @@ func ManageUser(c *gin.Context) {
 		return
 	}
 	myRole := c.GetInt("role")
-	if myRole <= user.Role && myRole != common.RoleRootUser {
+	if !canManageTargetRole(myRole, user.Role) {
 		common.ApiErrorI18n(c, i18n.MsgUserNoPermissionHigherLevel)
 		return
 	}

docker-compose.yml

@@ -34,6 +34,7 @@ services:
       - BATCH_UPDATE_ENABLED=true  # 是否启用批量更新 (Whether to enable batch update)
       - NODE_NAME=new-api-node-1  # 节点名称，用于审计日志中标识节点身份；多节点/容器部署时建议设置 (Node name used in audit logs; recommended when running multiple instances or in containers)
 #      - STREAMING_TIMEOUT=300  # 流模式无响应超时时间，单位秒，默认120秒，如果出现空补全可以尝试改为更大值 （Streaming timeout in seconds, default is 120s. Increase if experiencing empty completions）
+#      - RELAY_IDLE_CONN_TIMEOUT=90  # Relay HTTP 客户端空闲连接超时时间，单位秒，默认跟随 Go 标准库，设置为0表示不限制 (Relay HTTP client idle keep-alive timeout in seconds, defaults to Go standard library; set 0 to disable)
 #      - SESSION_SECRET=random_string  # 多机部署时设置，必须修改这个随机字符串！！ （multi-node deployment, set this to a random string!!!!!!!）
 #      - SYNC_FREQUENCY=60  # Uncomment if regular database syncing is needed
 #      - GOOGLE_ANALYTICS_ID=G-XXXXXXXXXX  # Google Analytics 的测量 ID (Google Analytics Measurement ID)

dto/openai_image.go

@@ -26,11 +26,11 @@ type ImageRequest struct {
 	OutputFormat      json.RawMessage `json:"output_format,omitempty"`
 	OutputCompression json.RawMessage `json:"output_compression,omitempty"`
 	PartialImages     json.RawMessage `json:"partial_images,omitempty"`
-	// Stream            bool            `json:"stream,omitempty"`
-	Images        json.RawMessage `json:"images,omitempty"`
-	Mask          json.RawMessage `json:"mask,omitempty"`
-	InputFidelity json.RawMessage `json:"input_fidelity,omitempty"`
-	Watermark     *bool           `json:"watermark,omitempty"`
+	Stream            *bool           `json:"stream,omitempty"`
+	Images            json.RawMessage `json:"images,omitempty"`
+	Mask              json.RawMessage `json:"mask,omitempty"`
+	InputFidelity     json.RawMessage `json:"input_fidelity,omitempty"`
+	Watermark         *bool           `json:"watermark,omitempty"`
 	// zhipu 4v
 	WatermarkEnabled json.RawMessage `json:"watermark_enabled,omitempty"`
 	UserId           json.RawMessage `json:"user_id,omitempty"`
@@ -163,7 +163,7 @@ func (i *ImageRequest) GetTokenCountMeta() *types.TokenCountMeta {
 }
 
 func (i *ImageRequest) IsStream(c *gin.Context) bool {
-	return false
+	return i.Stream != nil && *i.Stream
 }
 
 func (i *ImageRequest) SetModelName(modelName string) {

dto/openai_request.go

@@ -213,12 +213,22 @@ func (r *GeneralOpenAIRequest) ToMap() map[string]any {
 	return result
 }
 
+func IsOpenAIReasoningOModel(modelName string) bool {
+	return strings.HasPrefix(modelName, "o1") ||
+		strings.HasPrefix(modelName, "o3") ||
+		strings.HasPrefix(modelName, "o4")
+}
+
+func IsOpenAIGPT5Model(modelName string) bool {
+	return strings.HasPrefix(modelName, "gpt-5")
+}
+
 func (r *GeneralOpenAIRequest) GetSystemRoleName() string {
-	if strings.HasPrefix(r.Model, "o") {
+	if IsOpenAIReasoningOModel(r.Model) {
 		if !strings.HasPrefix(r.Model, "o1-mini") && !strings.HasPrefix(r.Model, "o1-preview") {
 			return "developer"
 		}
-	} else if strings.HasPrefix(r.Model, "gpt-5") {
+	} else if IsOpenAIGPT5Model(r.Model) {
 		return "developer"
 	}
 	return "system"

dto/openai_request_zero_value_test.go

@@ -71,3 +71,27 @@ func TestOpenAIResponsesRequestPreserveExplicitZeroValues(t *testing.T) {
 	require.True(t, gjson.GetBytes(encoded, "stream").Exists())
 	require.True(t, gjson.GetBytes(encoded, "top_p").Exists())
 }
+
+func TestGeneralOpenAIRequestGetSystemRoleName(t *testing.T) {
+	tests := []struct {
+		name  string
+		model string
+		want  string
+	}{
+		{name: "o1 uses developer", model: "o1", want: "developer"},
+		{name: "o3 family uses developer", model: "o3-mini-high", want: "developer"},
+		{name: "o4 family uses developer", model: "o4-mini", want: "developer"},
+		{name: "o1 mini stays system", model: "o1-mini", want: "system"},
+		{name: "o1 preview stays system", model: "o1-preview", want: "system"},
+		{name: "gpt 5 uses developer", model: "gpt-5", want: "developer"},
+		{name: "omni is not o series", model: "omni-moderation-latest", want: "system"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			req := GeneralOpenAIRequest{Model: tt.model}
+
+			require.Equal(t, tt.want, req.GetSystemRoleName())
+		})
+	}
+}

go.mod

@@ -60,6 +60,8 @@ require (
 	gorm.io/gorm v1.25.2
 )
 
+require github.com/waffo-com/waffo-pancake-sdk-go v0.3.1
+
 require (
 	github.com/DmitriyVTitov/size v1.5.0 // indirect
 	github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6 // indirect

go.sum

@@ -308,6 +308,12 @@ github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65E
 github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
 github.com/waffo-com/waffo-go v1.3.1 h1:NCYD3oQ59DTJj1bwS5T/659LI4h8PuAIW4Qj/w7fKPw=
 github.com/waffo-com/waffo-go v1.3.1/go.mod h1:IaXVYq6mmYtrLFFsLxPslNwuIZx0mIadWWjhe+eWb0g=
+github.com/waffo-com/waffo-pancake-sdk-go v0.1.1 h1:YOI7+3zTBlTB7Ou6+ZXnJV2JvW/ag9d7CwE/TxH3Hls=
+github.com/waffo-com/waffo-pancake-sdk-go v0.1.1/go.mod h1:5MBCGH/nqRRA5sHO/lQB/96r4BTAqy8QpWxn53m9htI=
+github.com/waffo-com/waffo-pancake-sdk-go v0.2.0 h1:cCSgccM66p7feTtgRqUUGT50tYQOhahsoPXavd+ib1U=
+github.com/waffo-com/waffo-pancake-sdk-go v0.2.0/go.mod h1:5MBCGH/nqRRA5sHO/lQB/96r4BTAqy8QpWxn53m9htI=
+github.com/waffo-com/waffo-pancake-sdk-go v0.3.1 h1:ngQSN/oVB35xTwFPLfg++bxPC+SptcF145Mb6c62YCc=
+github.com/waffo-com/waffo-pancake-sdk-go v0.3.1/go.mod h1:OB2MyFIQaefoPO0FV3J+yu9sDP8RVFQ+sbFsXqGuObc=
 github.com/x448/float16 v0.8.4 h1:qLwI1I70+NjRFUR3zs1JPUCgaCXSh3SW62uAKT1mSBM=
 github.com/x448/float16 v0.8.4/go.mod h1:14CWIYCyZA/cWjXOioeEpHeN/83MdbZDRQHoFcYsOfg=
 github.com/xyproto/randomstring v1.0.5 h1:YtlWPoRdgMu3NZtP45drfy1GKoojuR7hmRcnhZqKjWU=

logger/logger.go

@@ -95,9 +95,11 @@ func LogDebug(ctx context.Context, msg string, args ...any) {
 }
 
 func logHelper(ctx context.Context, level string, msg string) {
-	id := ctx.Value(common.RequestIdKey)
-	if id == nil {
-		id = "SYSTEM"
+	var id any = "SYSTEM"
+	if ctx != nil {
+		if requestID := ctx.Value(common.RequestIdKey); requestID != nil {
+			id = requestID
+		}
 	}
 	now := time.Now()
 	common.LogWriterMu.RLock()
@@ -172,10 +174,13 @@ func FormatQuota(quota int) string {
 
 // LogJson 仅供测试使用 only for test
 func LogJson(ctx context.Context, msg string, obj any) {
+	if !common.DebugEnabled {
+		return
+	}
 	jsonStr, err := common.Marshal(obj)
 	if err != nil {
 		LogError(ctx, fmt.Sprintf("json marshal failed: %s", err.Error()))
 		return
 	}
-	LogDebug(ctx, fmt.Sprintf("%s | %s", msg, string(jsonStr)))
+	LogDebug(ctx, "%s | %s", msg, jsonStr)
 }

makefile

@@ -1,18 +1,28 @@
 FRONTEND_DIR = ./web/default
 FRONTEND_CLASSIC_DIR = ./web/classic
 BACKEND_DIR = .
+DEV_FRONTEND_DEFAULT_PORT ?= 5173
+DEV_FRONTEND_CLASSIC_PORT ?= 5174
+DEV_COMPOSE_FILE = docker-compose.dev.yml
+DEV_POSTGRES_SERVICE = postgres
+DEV_BACKEND_SERVICE = new-api
+DEV_POSTGRES_DB = new-api
+DEV_POSTGRES_USER = root
+DEV_SQLITE_PATH ?= one-api.db
 
-.PHONY: all build-frontend build-frontend-classic build-all-frontends start-backend dev dev-api dev-web dev-web-classic
+.PHONY: all build-frontend build-frontend-classic build-all-frontends start-backend dev dev-api dev-api-rebuild dev-web dev-web-classic reset-setup
 
 all: build-all-frontends start-backend
 
 build-frontend:
 	@echo "Building default frontend..."
-	@cd $(FRONTEND_DIR) && bun install && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
+	@cd ./web && bun install --frozen-lockfile
+	@cd $(FRONTEND_DIR) && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
 
 build-frontend-classic:
 	@echo "Building classic frontend..."
-	@cd $(FRONTEND_CLASSIC_DIR) && bun install && VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
+	@cd ./web && bun install --frozen-lockfile
+	@cd $(FRONTEND_CLASSIC_DIR) && VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
 
 build-all-frontends: build-frontend build-frontend-classic
 
@@ -22,14 +32,65 @@ start-backend:
 
 dev-api:
 	@echo "Starting backend services (docker)..."
-	@docker compose -f docker-compose.dev.yml up -d
+	@docker compose -f $(DEV_COMPOSE_FILE) up -d
+
+dev-api-rebuild:
+	@echo "Rebuilding and starting backend service (docker)..."
+	@docker compose -f $(DEV_COMPOSE_FILE) up -d --build $(DEV_BACKEND_SERVICE)
 
 dev-web:
-	@echo "Starting frontend dev server..."
-	@cd $(FRONTEND_DIR) && bun install && bun run dev
+	@echo "Starting both frontend dev servers..."
+	@echo "Default frontend: http://localhost:$(DEV_FRONTEND_DEFAULT_PORT)"
+	@echo "Classic frontend: http://localhost:$(DEV_FRONTEND_CLASSIC_PORT)"
+	@cd ./web && bun install
+	@(cd $(FRONTEND_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_DEFAULT_PORT)) & \
+		default_pid=$$!; \
+		(cd $(FRONTEND_CLASSIC_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_CLASSIC_PORT)) & \
+		classic_pid=$$!; \
+		trap 'kill $$default_pid $$classic_pid 2>/dev/null; wait $$default_pid $$classic_pid 2>/dev/null; exit 130' INT TERM; \
+		while kill -0 $$default_pid 2>/dev/null && kill -0 $$classic_pid 2>/dev/null; do \
+			sleep 1; \
+		done; \
+		if ! kill -0 $$default_pid 2>/dev/null; then \
+			wait $$default_pid; \
+			status=$$?; \
+			kill $$classic_pid 2>/dev/null; \
+			wait $$classic_pid 2>/dev/null; \
+			exit $$status; \
+		fi; \
+		wait $$classic_pid; \
+		status=$$?; \
+		kill $$default_pid 2>/dev/null; \
+		wait $$default_pid 2>/dev/null; \
+		exit $$status
 
 dev-web-classic:
 	@echo "Starting classic frontend dev server..."
-	@cd $(FRONTEND_CLASSIC_DIR) && bun install && bun run dev
+	@cd ./web && bun install
+	@cd $(FRONTEND_CLASSIC_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_CLASSIC_PORT)
 
 dev: dev-api dev-web
+
+reset-setup:
+	@echo "Resetting local setup wizard state..."
+	@if docker compose -f $(DEV_COMPOSE_FILE) ps --services --status running | grep -qx "$(DEV_POSTGRES_SERVICE)"; then \
+		echo "Detected running docker dev PostgreSQL. Removing setup record and root users..."; \
+		docker compose -f $(DEV_COMPOSE_FILE) exec -T $(DEV_POSTGRES_SERVICE) \
+			psql -U $(DEV_POSTGRES_USER) -d $(DEV_POSTGRES_DB) \
+			-c 'DELETE FROM setups;' \
+			-c 'DELETE FROM users WHERE role = 100;' \
+			-c "DELETE FROM options WHERE key IN ('SelfUseModeEnabled', 'DemoSiteEnabled');"; \
+		echo "Restarting docker dev backend so setup status is recalculated..."; \
+		docker compose -f $(DEV_COMPOSE_FILE) restart $(DEV_BACKEND_SERVICE); \
+	elif db_path="$${SQLITE_PATH:-$(DEV_SQLITE_PATH)}"; db_path="$${db_path%%\?*}"; [ -f "$$db_path" ]; then \
+		db_path="$${SQLITE_PATH:-$(DEV_SQLITE_PATH)}"; \
+		db_path="$${db_path%%\?*}"; \
+		echo "Detected local SQLite database: $$db_path"; \
+		sqlite3 "$$db_path" \
+			"DELETE FROM setups; DELETE FROM users WHERE role = 100; DELETE FROM options WHERE key IN ('SelfUseModeEnabled', 'DemoSiteEnabled');"; \
+		echo "SQLite setup state reset. Restart the local backend process before testing the setup wizard."; \
+	else \
+		echo "No running docker dev PostgreSQL or local SQLite database found."; \
+		echo "Start the dev stack with 'make dev-api', or set SQLITE_PATH/DEV_SQLITE_PATH to your local SQLite database."; \
+		exit 1; \
+	fi

middleware/distributor.go

@@ -3,6 +3,7 @@ package middleware
 import (
 	"errors"
 	"fmt"
+	"io"
 	"net/http"
 	"slices"
 	"strconv"
@@ -20,6 +21,7 @@ import (
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/gin-gonic/gin"
+	"github.com/tidwall/gjson"
 )
 
 type ModelRequest struct {
@@ -100,14 +102,10 @@ func Distribute() func(c *gin.Context) {
 				}
 
 				if preferredChannelID, found := service.GetPreferredChannelByAffinity(c, modelRequest.Model, usingGroup); found {
+					affinityUsable := false
 					preferred, err := model.CacheGetChannel(preferredChannelID)
-					if err == nil && preferred != nil {
-						if preferred.Status != common.ChannelStatusEnabled {
-							if service.ShouldSkipRetryAfterChannelAffinityFailure(c) {
-								abortWithOpenAiMessage(c, http.StatusForbidden, i18n.T(c, i18n.MsgDistributorAffinityChannelDisabled))
-								return
-							}
-						} else if usingGroup == "auto" {
+					if err == nil && preferred != nil && preferred.Status == common.ChannelStatusEnabled {
+						if usingGroup == "auto" {
 							userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
 							autoGroups := service.GetUserAutoGroup(userGroup)
 							for _, g := range autoGroups {
@@ -115,6 +113,7 @@ func Distribute() func(c *gin.Context) {
 									selectGroup = g
 									common.SetContextKey(c, constant.ContextKeyAutoGroup, g)
 									channel = preferred
+									affinityUsable = true
 									service.MarkChannelAffinityUsed(c, g, preferred.Id)
 									break
 								}
@@ -122,9 +121,13 @@ func Distribute() func(c *gin.Context) {
 						} else if model.IsChannelEnabledForGroupModel(usingGroup, modelRequest.Model, preferred.Id) {
 							channel = preferred
 							selectGroup = usingGroup
+							affinityUsable = true
 							service.MarkChannelAffinityUsed(c, usingGroup, preferred.Id)
 						}
 					}
+					if !affinityUsable && !service.ShouldKeepChannelAffinityOnChannelDisabled() {
+						service.ClearCurrentChannelAffinityCache(c)
+					}
 				}
 
 				if channel == nil {
@@ -170,6 +173,14 @@ func Distribute() func(c *gin.Context) {
 // - application/x-www-form-urlencoded
 // - multipart/form-data
 func getModelFromRequest(c *gin.Context) (*ModelRequest, error) {
+	if strings.HasPrefix(c.Request.Header.Get("Content-Type"), "application/json") {
+		modelRequest, err := getModelFromJSONBody(c)
+		if err != nil {
+			return nil, errors.New(i18n.T(c, i18n.MsgDistributorInvalidRequest, map[string]any{"Error": err.Error()}))
+		}
+		return modelRequest, nil
+	}
+
 	var modelRequest ModelRequest
 	err := common.UnmarshalBodyReusable(c, &modelRequest)
 	if err != nil {
@@ -178,6 +189,50 @@ func getModelFromRequest(c *gin.Context) (*ModelRequest, error) {
 	return &modelRequest, nil
 }
 
+func getModelFromJSONBody(c *gin.Context) (*ModelRequest, error) {
+	storage, err := common.GetBodyStorage(c)
+	if err != nil {
+		return nil, err
+	}
+	requestBody, err := storage.Bytes()
+	if err != nil {
+		return nil, err
+	}
+	if !gjson.ValidBytes(requestBody) {
+		return nil, errors.New("invalid JSON request body")
+	}
+
+	values := gjson.GetManyBytes(requestBody, "model", "group")
+	model, err := getJSONStringValue(values[0], "model")
+	if err != nil {
+		return nil, err
+	}
+	group, err := getJSONStringValue(values[1], "group")
+	if err != nil {
+		return nil, err
+	}
+
+	if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
+		return nil, seekErr
+	}
+	c.Request.Body = io.NopCloser(storage)
+
+	return &ModelRequest{
+		Model: model,
+		Group: group,
+	}, nil
+}
+
+func getJSONStringValue(result gjson.Result, field string) (string, error) {
+	if !result.Exists() || result.Type == gjson.Null {
+		return "", nil
+	}
+	if result.Type != gjson.String {
+		return "", fmt.Errorf("field %s must be a string", field)
+	}
+	return result.String(), nil
+}
+
 func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 	var modelRequest ModelRequest
 	shouldSelectChannel := true
@@ -244,6 +299,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 		} else if c.Request.Method == http.MethodGet {
 			relayMode = relayconstant.RelayModeVideoFetchByID
 			shouldSelectChannel = false
+			modelRequest.Model = getTaskOriginModelName(c)
 		}
 		c.Set("relay_mode", relayMode)
 	} else if strings.Contains(c.Request.URL.Path, "/v1/video/generations") {
@@ -258,6 +314,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 		} else if c.Request.Method == http.MethodGet {
 			relayMode = relayconstant.RelayModeVideoFetchByID
 			shouldSelectChannel = false
+			modelRequest.Model = getTaskOriginModelName(c)
 		}
 		if _, ok := c.Get("relay_mode"); !ok {
 			c.Set("relay_mode", relayMode)
@@ -342,6 +399,31 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 	return &modelRequest, shouldSelectChannel, nil
 }
 
+// 修复 #4834: GET /v1/video/generations/:task_id && /v1/video/:task_id 此前不解析 model，
+// 当 token 启用「可用模型限制」时，下游 modelLimitEnable 校验会因
+// modelRequest.Model 为空而误报 "This token has no access to model"。
+// 从已存储的任务记录中回填 OriginModelName 即可让校验走在正确的模型上。
+func getTaskOriginModelName(c *gin.Context) string {
+	if !common.GetContextKeyBool(c, constant.ContextKeyTokenModelLimitEnabled) {
+		return ""
+	}
+
+	taskId := c.Param("task_id")
+	if taskId == "" {
+		// jimeng adapter
+		taskId = c.GetString("task_id")
+	}
+	if taskId == "" {
+		return ""
+	}
+
+	userId := c.GetInt("id")
+	if task, exist, err := model.GetByTaskId(userId, taskId); err == nil && exist && task != nil {
+		return task.Properties.OriginModelName
+	}
+	return ""
+}
+
 func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, modelName string) *types.NewAPIError {
 	c.Set("original_model", modelName) // for retry
 	if channel == nil {

middleware/header_nav.go

@@ -0,0 +1,135 @@
+package middleware
+
+import (
+	"fmt"
+	"net/http"
+	"strings"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/gin-gonic/gin"
+)
+
+type headerNavAccess struct {
+	Enabled     bool
+	RequireAuth bool
+}
+
+func getHeaderNavAccess(module string) headerNavAccess {
+	fallback := headerNavAccess{
+		Enabled:     true,
+		RequireAuth: false,
+	}
+
+	common.OptionMapRWMutex.RLock()
+	raw := common.OptionMap["HeaderNavModules"]
+	common.OptionMapRWMutex.RUnlock()
+
+	if strings.TrimSpace(raw) == "" {
+		return fallback
+	}
+
+	var parsed map[string]any
+	if err := common.Unmarshal([]byte(raw), &parsed); err != nil {
+		return fallback
+	}
+
+	return parseHeaderNavAccess(parsed[module], fallback)
+}
+
+func parseHeaderNavAccess(raw any, fallback headerNavAccess) headerNavAccess {
+	switch value := raw.(type) {
+	case bool:
+		return headerNavAccess{
+			Enabled:     value,
+			RequireAuth: fallback.RequireAuth,
+		}
+	case string:
+		return headerNavAccess{
+			Enabled:     parseHeaderNavBool(value, fallback.Enabled),
+			RequireAuth: fallback.RequireAuth,
+		}
+	case float64:
+		return headerNavAccess{
+			Enabled:     parseHeaderNavBool(value, fallback.Enabled),
+			RequireAuth: fallback.RequireAuth,
+		}
+	case map[string]any:
+		access := fallback
+		if enabled, ok := value["enabled"]; ok {
+			access.Enabled = parseHeaderNavBool(enabled, fallback.Enabled)
+		}
+		if requireAuth, ok := value["requireAuth"]; ok {
+			access.RequireAuth = parseHeaderNavBool(requireAuth, fallback.RequireAuth)
+		}
+		return access
+	default:
+		return fallback
+	}
+}
+
+func parseHeaderNavBool(value any, fallback bool) bool {
+	switch v := value.(type) {
+	case bool:
+		return v
+	case string:
+		switch strings.ToLower(strings.TrimSpace(v)) {
+		case "true", "1":
+			return true
+		case "false", "0":
+			return false
+		default:
+			return fallback
+		}
+	case float64:
+		if v == 1 {
+			return true
+		}
+		if v == 0 {
+			return false
+		}
+		return fallback
+	case int:
+		if v == 1 {
+			return true
+		}
+		if v == 0 {
+			return false
+		}
+		return fallback
+	default:
+		return fallback
+	}
+}
+
+func HeaderNavModuleAuth(module string) gin.HandlerFunc {
+	return func(c *gin.Context) {
+		access := getHeaderNavAccess(module)
+		if !access.Enabled {
+			c.JSON(http.StatusForbidden, gin.H{
+				"success": false,
+				"message": fmt.Sprintf("%s is disabled", module),
+			})
+			c.Abort()
+			return
+		}
+
+		if access.RequireAuth {
+			UserAuth()(c)
+			return
+		}
+
+		TryUserAuth()(c)
+	}
+}
+
+func HeaderNavModulePublicOrUserAuth(module string) gin.HandlerFunc {
+	return func(c *gin.Context) {
+		access := getHeaderNavAccess(module)
+		if !access.Enabled || access.RequireAuth {
+			UserAuth()(c)
+			return
+		}
+
+		TryUserAuth()(c)
+	}
+}

middleware/header_nav_test.go

@@ -0,0 +1,167 @@
+package middleware
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/gin-contrib/sessions"
+	"github.com/gin-contrib/sessions/cookie"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+func withHeaderNavModules(t *testing.T, raw string) {
+	t.Helper()
+
+	common.OptionMapRWMutex.Lock()
+	if common.OptionMap == nil {
+		common.OptionMap = map[string]string{}
+	}
+	previous, hadPrevious := common.OptionMap["HeaderNavModules"]
+	common.OptionMap["HeaderNavModules"] = raw
+	common.OptionMapRWMutex.Unlock()
+
+	t.Cleanup(func() {
+		common.OptionMapRWMutex.Lock()
+		defer common.OptionMapRWMutex.Unlock()
+		if hadPrevious {
+			common.OptionMap["HeaderNavModules"] = previous
+			return
+		}
+		delete(common.OptionMap, "HeaderNavModules")
+	})
+}
+
+func performHeaderNavRequest(t *testing.T, handler gin.HandlerFunc, authenticated bool) *httptest.ResponseRecorder {
+	t.Helper()
+
+	gin.SetMode(gin.TestMode)
+	router := gin.New()
+	router.Use(sessions.Sessions("session", cookie.NewStore([]byte("header-nav-test"))))
+	router.GET("/login", func(c *gin.Context) {
+		session := sessions.Default(c)
+		session.Set("username", "tester")
+		session.Set("role", common.RoleCommonUser)
+		session.Set("id", 1)
+		session.Set("status", common.UserStatusEnabled)
+		session.Set("group", "default")
+		if err := session.Save(); err != nil {
+			c.JSON(http.StatusInternalServerError, gin.H{"success": false})
+			return
+		}
+		c.Status(http.StatusNoContent)
+	})
+	router.GET("/api/test", handler, func(c *gin.Context) {
+		c.JSON(http.StatusOK, gin.H{"success": true})
+	})
+
+	var cookies []*http.Cookie
+	if authenticated {
+		loginRecorder := httptest.NewRecorder()
+		loginRequest := httptest.NewRequest(http.MethodGet, "/login", nil)
+		router.ServeHTTP(loginRecorder, loginRequest)
+		require.Equal(t, http.StatusNoContent, loginRecorder.Code)
+		cookies = loginRecorder.Result().Cookies()
+	}
+
+	recorder := httptest.NewRecorder()
+	request := httptest.NewRequest(http.MethodGet, "/api/test", nil)
+	if authenticated {
+		request.Header.Set("New-Api-User", "1")
+		for _, cookie := range cookies {
+			request.AddCookie(cookie)
+		}
+	}
+	router.ServeHTTP(recorder, request)
+	return recorder
+}
+
+func TestHeaderNavModuleAuthAllowsDefaultPublicAccess(t *testing.T) {
+	withHeaderNavModules(t, "")
+
+	recorder := performHeaderNavRequest(t, HeaderNavModuleAuth("pricing"), false)
+
+	require.Equal(t, http.StatusOK, recorder.Code)
+}
+
+func TestHeaderNavModuleAuthRejectsDisabledPricing(t *testing.T) {
+	raw := `{"pricing":{"enabled":false,"requireAuth":false}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModuleAuth("pricing"), false)
+
+	require.Equal(t, http.StatusForbidden, recorder.Code)
+}
+
+func TestHeaderNavModuleAuthRequiresLoginForPricing(t *testing.T) {
+	raw := `{"pricing":{"enabled":true,"requireAuth":true}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModuleAuth("pricing"), false)
+
+	require.Equal(t, http.StatusUnauthorized, recorder.Code)
+}
+
+func TestHeaderNavModuleAuthRequiresLoginForRankings(t *testing.T) {
+	raw := `{"rankings":{"enabled":true,"requireAuth":true}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModuleAuth("rankings"), false)
+
+	require.Equal(t, http.StatusUnauthorized, recorder.Code)
+}
+
+func TestHeaderNavModuleAuthRejectsLegacyDisabledModule(t *testing.T) {
+	raw := `{"rankings":false}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModuleAuth("rankings"), false)
+
+	require.Equal(t, http.StatusForbidden, recorder.Code)
+}
+
+func TestHeaderNavModulePublicOrUserAuthAllowsDefaultPublicAccess(t *testing.T) {
+	withHeaderNavModules(t, "")
+
+	recorder := performHeaderNavRequest(t, HeaderNavModulePublicOrUserAuth("pricing"), false)
+
+	require.Equal(t, http.StatusOK, recorder.Code)
+}
+
+func TestHeaderNavModulePublicOrUserAuthRequiresLoginWhenDisabled(t *testing.T) {
+	raw := `{"pricing":{"enabled":false,"requireAuth":false}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModulePublicOrUserAuth("pricing"), false)
+
+	require.Equal(t, http.StatusUnauthorized, recorder.Code)
+}
+
+func TestHeaderNavModulePublicOrUserAuthAllowsLoggedInWhenDisabled(t *testing.T) {
+	raw := `{"pricing":{"enabled":false,"requireAuth":false}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModulePublicOrUserAuth("pricing"), true)
+
+	require.Equal(t, http.StatusOK, recorder.Code)
+}
+
+func TestHeaderNavModulePublicOrUserAuthRequiresLoginWhenRequireAuth(t *testing.T) {
+	raw := `{"pricing":{"enabled":true,"requireAuth":true}}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModulePublicOrUserAuth("pricing"), false)
+
+	require.Equal(t, http.StatusUnauthorized, recorder.Code)
+}
+
+func TestHeaderNavModulePublicOrUserAuthRequiresLoginForLegacyDisabledModule(t *testing.T) {
+	raw := `{"pricing":false}`
+	withHeaderNavModules(t, raw)
+
+	recorder := performHeaderNavRequest(t, HeaderNavModulePublicOrUserAuth("pricing"), false)
+
+	require.Equal(t, http.StatusUnauthorized, recorder.Code)
+}

middleware/request_body_limit.go

@@ -0,0 +1,47 @@
+package middleware
+
+import (
+	"bytes"
+	"io"
+	"net/http"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/gin-gonic/gin"
+)
+
+func AnonymousRequestBodyLimit() gin.HandlerFunc {
+	return func(c *gin.Context) {
+		maxBytes := common.GetAnonymousRequestBodyLimitBytes()
+		if maxBytes <= 0 || c.Request.Body == nil {
+			c.Next()
+			return
+		}
+
+		originalBody := c.Request.Body
+		limitedBody, err := readAnonymousRequestBody(originalBody, maxBytes)
+		_ = originalBody.Close()
+		if err != nil {
+			if common.IsRequestBodyTooLargeError(err) {
+				c.AbortWithStatus(http.StatusRequestEntityTooLarge)
+				return
+			}
+			c.AbortWithStatus(http.StatusBadRequest)
+			return
+		}
+
+		c.Request.Body = io.NopCloser(bytes.NewReader(limitedBody))
+		c.Request.ContentLength = int64(len(limitedBody))
+		c.Next()
+	}
+}
+
+func readAnonymousRequestBody(body io.Reader, maxBytes int64) ([]byte, error) {
+	data, err := io.ReadAll(io.LimitReader(body, maxBytes+1))
+	if err != nil {
+		return nil, err
+	}
+	if int64(len(data)) > maxBytes {
+		return nil, common.ErrRequestBodyTooLarge
+	}
+	return data, nil
+}

model/channel.go

@@ -12,6 +12,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/types"
 
 	"github.com/samber/lo"
@@ -128,6 +129,38 @@ func resolveChannelSortOptions(idSort bool, sortOptions []ChannelSortOptions) Ch
 	return options
 }
 
+func NormalizeChannelGroupFilter(group string) string {
+	group = strings.TrimSpace(group)
+	if group == "" || strings.EqualFold(group, "all") || strings.EqualFold(group, "null") {
+		return ""
+	}
+	return group
+}
+
+func channelGroupFilterCondition() string {
+	if common.UsingMySQL {
+		return `CONCAT(',', ` + commonGroupCol + `, ',') LIKE ? ESCAPE '!'`
+	}
+	return `(',' || ` + commonGroupCol + ` || ',') LIKE ? ESCAPE '!'`
+}
+
+func channelGroupFilterPattern(group string) string {
+	group = strings.NewReplacer(
+		"!", "!!",
+		"%", "!%",
+		"_", "!_",
+	).Replace(group)
+	return "%," + group + ",%"
+}
+
+func ApplyChannelGroupFilter(query *gorm.DB, group string) *gorm.DB {
+	group = NormalizeChannelGroupFilter(group)
+	if group == "" {
+		return query
+	}
+	return query.Where(channelGroupFilterCondition(), channelGroupFilterPattern(group))
+}
+
 // Value implements driver.Valuer interface
 func (c ChannelInfo) Value() (driver.Value, error) {
 	return common.Marshal(&c)
@@ -218,10 +251,9 @@ func (channel *Channel) GetNextEnabledKey() (string, int, *types.NewAPIError) {
 		if err != nil {
 			return "", 0, types.NewError(err, types.ErrorCodeGetChannelFailed, types.ErrOptionWithSkipRetry())
 		}
-		//println("before polling index:", channel.ChannelInfo.MultiKeyPollingIndex)
 		defer func() {
 			if common.DebugEnabled {
-				println(fmt.Sprintf("channel %d polling index: %d", channel.Id, channel.ChannelInfo.MultiKeyPollingIndex))
+				logger.LogDebug(nil, "channel %d polling index: %d", channel.Id, channel.ChannelInfo.MultiKeyPollingIndex)
 			}
 			if !common.MemoryCacheEnabled {
 				_ = channel.SaveChannelInfo()
@@ -365,25 +397,12 @@ func SearchChannels(keyword string, group string, model string, idSort bool, sor
 	baseQuery := DB.Model(&Channel{}).Omit("key")
 
 	// 构造WHERE子句
-	var whereClause string
-	var args []interface{}
-	if group != "" && group != "null" {
-		var groupCondition string
-		if common.UsingMySQL {
-			groupCondition = `CONCAT(',', ` + commonGroupCol + `, ',') LIKE ?`
-		} else {
-			// sqlite, PostgreSQL
-			groupCondition = `(',' || ` + commonGroupCol + ` || ',') LIKE ?`
-		}
-		whereClause = "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + ` LIKE ? AND ` + groupCondition
-		args = append(args, common.String2Int(keyword), "%"+keyword+"%", keyword, "%"+keyword+"%", "%"+model+"%", "%,"+group+",%")
-	} else {
-		whereClause = "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + " LIKE ?"
-		args = append(args, common.String2Int(keyword), "%"+keyword+"%", keyword, "%"+keyword+"%", "%"+model+"%")
-	}
+	whereClause := "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + " LIKE ?"
+	args := []any{common.String2Int(keyword), "%" + keyword + "%", keyword, "%" + keyword + "%", "%" + model + "%"}
+	baseQuery = ApplyChannelGroupFilter(baseQuery.Where(whereClause, args...), group)
 
 	// 执行查询
-	err := order.Apply(baseQuery.Where(whereClause, args...)).Find(&channels).Error
+	err := order.Apply(baseQuery).Find(&channels).Error
 	if err != nil {
 		return nil, err
 	}
@@ -624,13 +643,25 @@ func handlerMultiKeyUpdate(channel *Channel, usingKey string, status int, reason
 	if len(keys) == 0 {
 		channel.Status = status
 	} else {
-		var keyIndex int
+		keyIndex := -1
 		for i, key := range keys {
 			if key == usingKey {
 				keyIndex = i
 				break
 			}
 		}
+		if keyIndex < 0 {
+			if usingKey != "" {
+				common.SysLog(fmt.Sprintf("failed to update multi-key status: channel_id=%d, using key not found", channel.Id))
+				return
+			}
+			channel.Status = status
+			info := channel.GetOtherInfo()
+			info["status_reason"] = reason
+			info["status_time"] = common.GetTimestamp()
+			channel.SetOtherInfo(info)
+			return
+		}
 		if channel.ChannelInfo.MultiKeyStatusList == nil {
 			channel.ChannelInfo.MultiKeyStatusList = make(map[int]int)
 		}
@@ -647,16 +678,31 @@ func handlerMultiKeyUpdate(channel *Channel, usingKey string, status int, reason
 			channel.ChannelInfo.MultiKeyDisabledReason[keyIndex] = reason
 			channel.ChannelInfo.MultiKeyDisabledTime[keyIndex] = common.GetTimestamp()
 		}
-		if len(channel.ChannelInfo.MultiKeyStatusList) >= channel.ChannelInfo.MultiKeySize {
+		if !hasEnabledMultiKey(keys, channel.ChannelInfo.MultiKeyStatusList) {
 			channel.Status = common.ChannelStatusAutoDisabled
 			info := channel.GetOtherInfo()
 			info["status_reason"] = "All keys are disabled"
 			info["status_time"] = common.GetTimestamp()
 			channel.SetOtherInfo(info)
+		} else if status == common.ChannelStatusEnabled {
+			channel.Status = common.ChannelStatusEnabled
 		}
 	}
 }
 
+func hasEnabledMultiKey(keys []string, statusList map[int]int) bool {
+	for i := range keys {
+		if statusList == nil {
+			return true
+		}
+		status, ok := statusList[i]
+		if !ok || status == common.ChannelStatusEnabled {
+			return true
+		}
+	}
+	return false
+}
+
 func UpdateChannelStatus(channelId int, usingKey string, status int, reason string) bool {
 	if common.MemoryCacheEnabled {
 		channelStatusLock.Lock()
@@ -668,11 +714,15 @@ func UpdateChannelStatus(channelId int, usingKey string, status int, reason stri
 		}
 		if channelCache.ChannelInfo.IsMultiKey {
 			// Use per-channel lock to prevent concurrent map read/write with GetNextEnabledKey
+			beforeStatus := channelCache.Status
 			pollingLock := GetChannelPollingLock(channelId)
 			pollingLock.Lock()
 			// 如果是多Key模式，更新缓存中的状态
 			handlerMultiKeyUpdate(channelCache, usingKey, status, reason)
 			pollingLock.Unlock()
+			if beforeStatus != channelCache.Status {
+				CacheUpdateChannelStatus(channelId, channelCache.Status)
+			}
 			//CacheUpdateChannel(channelCache)
 			//return true
 		} else {
@@ -828,8 +878,18 @@ func DeleteDisabledChannel() (int64, error) {
 }
 
 func GetPaginatedTags(offset int, limit int) ([]*string, error) {
+	return GetPaginatedChannelTags(DB.Model(&Channel{}), offset, limit)
+}
+
+func GetPaginatedChannelTags(query *gorm.DB, offset int, limit int) ([]*string, error) {
 	var tags []*string
-	err := DB.Model(&Channel{}).Select("DISTINCT tag").Where("tag != ''").Offset(offset).Limit(limit).Find(&tags).Error
+	err := query.
+		Select("DISTINCT tag").
+		Where("tag is not null AND tag != ''").
+		Order(clause.OrderByColumn{Column: clause.Column{Name: "tag"}}).
+		Offset(offset).
+		Limit(limit).
+		Find(&tags).Error
 	return tags, err
 }
 
@@ -857,24 +917,11 @@ func SearchTags(keyword string, group string, model string, idSort bool) ([]*str
 	baseQuery := DB.Model(&Channel{}).Omit("key")
 
 	// 构造WHERE子句
-	var whereClause string
-	var args []interface{}
-	if group != "" && group != "null" {
-		var groupCondition string
-		if common.UsingMySQL {
-			groupCondition = `CONCAT(',', ` + commonGroupCol + `, ',') LIKE ?`
-		} else {
-			// sqlite, PostgreSQL
-			groupCondition = `(',' || ` + commonGroupCol + ` || ',') LIKE ?`
-		}
-		whereClause = "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + ` LIKE ? AND ` + groupCondition
-		args = append(args, common.String2Int(keyword), "%"+keyword+"%", keyword, "%"+keyword+"%", "%"+model+"%", "%,"+group+",%")
-	} else {
-		whereClause = "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + " LIKE ?"
-		args = append(args, common.String2Int(keyword), "%"+keyword+"%", keyword, "%"+keyword+"%", "%"+model+"%")
-	}
+	whereClause := "(id = ? OR name LIKE ? OR " + commonKeyCol + " = ? OR " + baseURLCol + " LIKE ?) AND " + modelsCol + " LIKE ?"
+	args := []any{common.String2Int(keyword), "%" + keyword + "%", keyword, "%" + keyword + "%", "%" + model + "%"}
+	baseQuery = ApplyChannelGroupFilter(baseQuery.Where(whereClause, args...), group)
 
-	subQuery := baseQuery.Where(whereClause, args...).
+	subQuery := baseQuery.
 		Select("tag").
 		Where("tag != ''").
 		Order(order)
@@ -1015,8 +1062,12 @@ func CountAllChannels() (int64, error) {
 
 // CountAllTags returns number of non-empty distinct tags
 func CountAllTags() (int64, error) {
+	return CountChannelTags(DB.Model(&Channel{}))
+}
+
+func CountChannelTags(query *gorm.DB) (int64, error) {
 	var total int64
-	err := DB.Model(&Channel{}).Where("tag is not null AND tag != ''").Distinct("tag").Count(&total).Error
+	err := query.Where("tag is not null AND tag != ''").Distinct("tag").Count(&total).Error
 	return total, err
 }
 

model/channel_cache.go

@@ -11,6 +11,7 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
 )
 
@@ -257,9 +258,12 @@ func CacheUpdateChannel(channel *Channel) {
 		return
 	}
 
-	println("CacheUpdateChannel:", channel.Id, channel.Name, channel.Status, channel.ChannelInfo.MultiKeyPollingIndex)
-
-	println("before:", channelsIDM[channel.Id].ChannelInfo.MultiKeyPollingIndex)
+	if channelsIDM == nil {
+		channelsIDM = make(map[int]*Channel)
+	}
+	if oldChannel, ok := channelsIDM[channel.Id]; ok {
+		logger.LogDebug(nil, "CacheUpdateChannel before: id=%d, name=%s, status=%d, polling_index=%d", channel.Id, channel.Name, channel.Status, oldChannel.ChannelInfo.MultiKeyPollingIndex)
+	}
 	channelsIDM[channel.Id] = channel
-	println("after :", channelsIDM[channel.Id].ChannelInfo.MultiKeyPollingIndex)
+	logger.LogDebug(nil, "CacheUpdateChannel after: id=%d, name=%s, status=%d, polling_index=%d", channel.Id, channel.Name, channel.Status, channel.ChannelInfo.MultiKeyPollingIndex)
 }

model/log.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"errors"
 	"fmt"
+	"strings"
 	"time"
 
 	"github.com/QuantumNous/new-api/common"
@@ -16,25 +17,39 @@ import (
 	"gorm.io/gorm"
 )
 
+func applyExplicitLogTextFilter(tx *gorm.DB, column string, value string) (*gorm.DB, error) {
+	if value == "" {
+		return tx, nil
+	}
+	if strings.Contains(value, "%") {
+		pattern, err := sanitizeLikePattern(value)
+		if err != nil {
+			return nil, err
+		}
+		return tx.Where(column+" LIKE ? ESCAPE '!'", pattern), nil
+	}
+	return tx.Where(column+" = ?", value), nil
+}
+
 type Log struct {
-	Id               int    `json:"id" gorm:"index:idx_created_at_id,priority:1;index:idx_user_id_id,priority:2"`
-	UserId           int    `json:"user_id" gorm:"index;index:idx_user_id_id,priority:1"`
-	CreatedAt        int64  `json:"created_at" gorm:"bigint;index:idx_created_at_id,priority:2;index:idx_created_at_type"`
-	Type             int    `json:"type" gorm:"index:idx_created_at_type"`
-	Content          string `json:"content"`
-	Username         string `json:"username" gorm:"index;index:index_username_model_name,priority:2;default:''"`
-	TokenName        string `json:"token_name" gorm:"index;default:''"`
-	ModelName        string `json:"model_name" gorm:"index;index:index_username_model_name,priority:1;default:''"`
-	Quota            int    `json:"quota" gorm:"default:0"`
-	PromptTokens     int    `json:"prompt_tokens" gorm:"default:0"`
-	CompletionTokens int    `json:"completion_tokens" gorm:"default:0"`
-	UseTime          int    `json:"use_time" gorm:"default:0"`
-	IsStream         bool   `json:"is_stream"`
-	ChannelId        int    `json:"channel" gorm:"index"`
-	ChannelName      string `json:"channel_name" gorm:"->"`
-	TokenId          int    `json:"token_id" gorm:"default:0;index"`
-	Group            string `json:"group" gorm:"index"`
-	Ip               string `json:"ip" gorm:"index;default:''"`
+	Id                int    `json:"id" gorm:"index:idx_created_at_id,priority:2;index:idx_user_id_id,priority:2"`
+	UserId            int    `json:"user_id" gorm:"index;index:idx_user_id_id,priority:1"`
+	CreatedAt         int64  `json:"created_at" gorm:"bigint;index:idx_created_at_id,priority:1;index:idx_created_at_type"`
+	Type              int    `json:"type" gorm:"index:idx_created_at_type"`
+	Content           string `json:"content"`
+	Username          string `json:"username" gorm:"index;index:index_username_model_name,priority:2;default:''"`
+	TokenName         string `json:"token_name" gorm:"index;default:''"`
+	ModelName         string `json:"model_name" gorm:"index;index:index_username_model_name,priority:1;default:''"`
+	Quota             int    `json:"quota" gorm:"default:0"`
+	PromptTokens      int    `json:"prompt_tokens" gorm:"default:0"`
+	CompletionTokens  int    `json:"completion_tokens" gorm:"default:0"`
+	UseTime           int    `json:"use_time" gorm:"default:0"`
+	IsStream          bool   `json:"is_stream"`
+	ChannelId         int    `json:"channel" gorm:"index"`
+	ChannelName       string `json:"channel_name" gorm:"->"`
+	TokenId           int    `json:"token_id" gorm:"default:0;index"`
+	Group             string `json:"group" gorm:"index"`
+	Ip                string `json:"ip" gorm:"index;default:''"`
 	RequestId         string `json:"request_id,omitempty" gorm:"type:varchar(64);index:idx_logs_request_id;default:''"`
 	UpstreamRequestId string `json:"upstream_request_id,omitempty" gorm:"type:varchar(128);index:idx_logs_upstream_request_id;default:''"`
 	Other             string `json:"other"`
@@ -145,7 +160,7 @@ func RecordTopupLog(userId int, content string, callerIp string, paymentMethod s
 
 func RecordErrorLog(c *gin.Context, userId int, channelId int, modelName string, tokenName string, content string, tokenId int, useTimeSeconds int,
 	isStream bool, group string, other map[string]interface{}) {
-	logger.LogInfo(c, fmt.Sprintf("record error log: userId=%d, channelId=%d, modelName=%s, tokenName=%s, content=%s", userId, channelId, modelName, tokenName, content))
+	logger.LogInfo(c, fmt.Sprintf("record error log: userId=%d, channelId=%d, modelName=%s, tokenName=%s, content=%s", userId, channelId, modelName, tokenName, common.LocalLogPreview(content)))
 	username := c.GetString("username")
 	requestId := c.GetString(common.RequestIdKey)
 	upstreamRequestId := c.GetString(common.UpstreamRequestIdKey)
@@ -308,11 +323,11 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
 		tx = LOG_DB.Where("logs.type = ?", logType)
 	}
 
-	if modelName != "" {
-		tx = tx.Where("logs.model_name like ?", modelName)
+	if tx, err = applyExplicitLogTextFilter(tx, "logs.model_name", modelName); err != nil {
+		return nil, 0, err
 	}
-	if username != "" {
-		tx = tx.Where("logs.username = ?", username)
+	if tx, err = applyExplicitLogTextFilter(tx, "logs.username", username); err != nil {
+		return nil, 0, err
 	}
 	if tokenName != "" {
 		tx = tx.Where("logs.token_name = ?", tokenName)
@@ -339,7 +354,7 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
 	if err != nil {
 		return nil, 0, err
 	}
-	err = tx.Order("logs.id desc").Limit(num).Offset(startIdx).Find(&logs).Error
+	err = tx.Order("logs.created_at desc, logs.id desc").Limit(num).Offset(startIdx).Find(&logs).Error
 	if err != nil {
 		return nil, 0, err
 	}
@@ -397,12 +412,8 @@ func GetUserLogs(userId int, logType int, startTimestamp int64, endTimestamp int
 		tx = LOG_DB.Where("logs.user_id = ? and logs.type = ?", userId, logType)
 	}
 
-	if modelName != "" {
-		modelNamePattern, err := sanitizeLikePattern(modelName)
-		if err != nil {
-			return nil, 0, err
-		}
-		tx = tx.Where("logs.model_name LIKE ? ESCAPE '!'", modelNamePattern)
+	if tx, err = applyExplicitLogTextFilter(tx, "logs.model_name", modelName); err != nil {
+		return nil, 0, err
 	}
 	if tokenName != "" {
 		tx = tx.Where("logs.token_name = ?", tokenName)
@@ -449,9 +460,11 @@ func SumUsedQuota(logType int, startTimestamp int64, endTimestamp int64, modelNa
 	// 为rpm和tpm创建单独的查询
 	rpmTpmQuery := LOG_DB.Table("logs").Select("count(*) rpm, sum(prompt_tokens) + sum(completion_tokens) tpm")
 
-	if username != "" {
-		tx = tx.Where("username = ?", username)
-		rpmTpmQuery = rpmTpmQuery.Where("username = ?", username)
+	if tx, err = applyExplicitLogTextFilter(tx, "username", username); err != nil {
+		return stat, err
+	}
+	if rpmTpmQuery, err = applyExplicitLogTextFilter(rpmTpmQuery, "username", username); err != nil {
+		return stat, err
 	}
 	if tokenName != "" {
 		tx = tx.Where("token_name = ?", tokenName)
@@ -463,13 +476,11 @@ func SumUsedQuota(logType int, startTimestamp int64, endTimestamp int64, modelNa
 	if endTimestamp != 0 {
 		tx = tx.Where("created_at <= ?", endTimestamp)
 	}
-	if modelName != "" {
-		modelNamePattern, err := sanitizeLikePattern(modelName)
-		if err != nil {
-			return stat, err
-		}
-		tx = tx.Where("model_name LIKE ? ESCAPE '!'", modelNamePattern)
-		rpmTpmQuery = rpmTpmQuery.Where("model_name LIKE ? ESCAPE '!'", modelNamePattern)
+	if tx, err = applyExplicitLogTextFilter(tx, "model_name", modelName); err != nil {
+		return stat, err
+	}
+	if rpmTpmQuery, err = applyExplicitLogTextFilter(rpmTpmQuery, "model_name", modelName); err != nil {
+		return stat, err
 	}
 	if channel != 0 {
 		tx = tx.Where("channel_id = ?", channel)

model/main.go

@@ -397,8 +397,10 @@ func ensureSubscriptionPlanTableSQLite() error {
 ` + "`custom_seconds`" + ` bigint NOT NULL DEFAULT 0,
 ` + "`enabled`" + ` numeric DEFAULT 1,
 ` + "`sort_order`" + ` integer DEFAULT 0,
+` + "`allow_balance_pay`" + ` numeric DEFAULT 1,
 ` + "`stripe_price_id`" + ` varchar(128) DEFAULT '',
 ` + "`creem_product_id`" + ` varchar(128) DEFAULT '',
+` + "`waffo_pancake_product_id`" + ` varchar(128) DEFAULT '',
 ` + "`max_purchase_per_user`" + ` integer DEFAULT 0,
 ` + "`upgrade_group`" + ` varchar(64) DEFAULT '',
 ` + "`total_amount`" + ` bigint NOT NULL DEFAULT 0,
@@ -430,8 +432,10 @@ PRIMARY KEY (` + "`id`" + `)
 		{Name: "custom_seconds", DDL: "`custom_seconds` bigint NOT NULL DEFAULT 0"},
 		{Name: "enabled", DDL: "`enabled` numeric DEFAULT 1"},
 		{Name: "sort_order", DDL: "`sort_order` integer DEFAULT 0"},
+		{Name: "allow_balance_pay", DDL: "`allow_balance_pay` numeric DEFAULT 1"},
 		{Name: "stripe_price_id", DDL: "`stripe_price_id` varchar(128) DEFAULT ''"},
 		{Name: "creem_product_id", DDL: "`creem_product_id` varchar(128) DEFAULT ''"},
+		{Name: "waffo_pancake_product_id", DDL: "`waffo_pancake_product_id` varchar(128) DEFAULT ''"},
 		{Name: "max_purchase_per_user", DDL: "`max_purchase_per_user` integer DEFAULT 0"},
 		{Name: "upgrade_group", DDL: "`upgrade_group` varchar(64) DEFAULT ''"},
 		{Name: "total_amount", DDL: "`total_amount` bigint NOT NULL DEFAULT 0"},

model/model_meta.go

@@ -2,6 +2,7 @@ package model
 
 import (
 	"strconv"
+	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 
@@ -135,6 +136,62 @@ func GetBoundChannelsByModelsMap(modelNames []string) (map[string][]BoundChannel
 	return result, nil
 }
 
+func normalizeLookupValues(values []string) []string {
+	seen := make(map[string]struct{}, len(values))
+	normalized := make([]string, 0, len(values))
+	for _, value := range values {
+		value = strings.TrimSpace(value)
+		if value == "" {
+			continue
+		}
+		if _, ok := seen[value]; ok {
+			continue
+		}
+		seen[value] = struct{}{}
+		normalized = append(normalized, value)
+	}
+	return normalized
+}
+
+func GetPreferredModelOwnerChannelTypes(modelNames []string, groups []string) (map[string]int, error) {
+	result := make(map[string]int)
+	modelNames = normalizeLookupValues(modelNames)
+	if len(modelNames) == 0 {
+		return result, nil
+	}
+
+	type row struct {
+		Model       string
+		ChannelType int
+	}
+	var rows []row
+
+	query := DB.Table("abilities").
+		Select("abilities.model as model, channels.type as channel_type").
+		Joins("JOIN channels ON abilities.channel_id = channels.id").
+		Where("abilities.model IN ? AND abilities.enabled = ? AND channels.status = ?", modelNames, true, common.ChannelStatusEnabled).
+		Order("COALESCE(abilities.priority, 0) DESC").
+		Order("abilities.weight DESC").
+		Order("abilities.channel_id ASC")
+
+	groups = normalizeLookupValues(groups)
+	if len(groups) > 0 {
+		query = query.Where("abilities."+commonGroupCol+" IN ?", groups)
+	}
+
+	if err := query.Scan(&rows).Error; err != nil {
+		return nil, err
+	}
+
+	for _, r := range rows {
+		if _, ok := result[r.Model]; ok {
+			continue
+		}
+		result[r.Model] = r.ChannelType
+	}
+	return result, nil
+}
+
 func SearchModels(keyword string, vendor string, offset int, limit int) ([]*Model, int64, error) {
 	var models []*Model
 	db := DB.Model(&Model{})

model/model_owner_test.go

@@ -0,0 +1,141 @@
+package model
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/constant"
+	"github.com/stretchr/testify/require"
+)
+
+func clearPreferredOwnerTables(t *testing.T) {
+	t.Helper()
+	require.NoError(t, DB.Exec("DELETE FROM abilities").Error)
+	require.NoError(t, DB.Exec("DELETE FROM channels").Error)
+}
+
+func insertPreferredOwnerCandidate(
+	t *testing.T,
+	channelID int,
+	modelName string,
+	group string,
+	channelType int,
+	priority int64,
+	weight uint,
+	channelStatus int,
+	abilityEnabled bool,
+) {
+	t.Helper()
+	require.NoError(t, DB.Create(&Channel{
+		Id:     channelID,
+		Type:   channelType,
+		Key:    fmt.Sprintf("key-%d", channelID),
+		Status: channelStatus,
+		Name:   fmt.Sprintf("channel-%d", channelID),
+	}).Error)
+	require.NoError(t, DB.Create(&Ability{
+		Group:     group,
+		Model:     modelName,
+		ChannelId: channelID,
+		Enabled:   abilityEnabled,
+		Priority:  &priority,
+		Weight:    weight,
+	}).Error)
+}
+
+func TestGetPreferredModelOwnerChannelTypes(t *testing.T) {
+	const modelName = "gpt-5.4"
+
+	tests := []struct {
+		name     string
+		setup    func(t *testing.T)
+		groups   []string
+		expected int
+		found    bool
+	}{
+		{
+			name: "openai only",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 0, 0, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeOpenAI,
+			found:    true,
+		},
+		{
+			name: "codex only",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeCodex, 0, 0, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeCodex,
+			found:    true,
+		},
+		{
+			name: "priority wins",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 100, common.ChannelStatusEnabled, true)
+				insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 2, 0, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeCodex,
+			found:    true,
+		},
+		{
+			name: "weight wins when priority is equal",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 10, common.ChannelStatusEnabled, true)
+				insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 1, 20, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeCodex,
+			found:    true,
+		},
+		{
+			name: "channel id stabilizes exact ties",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 1, 10, common.ChannelStatusEnabled, true)
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 10, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeOpenAI,
+			found:    true,
+		},
+		{
+			name: "group filter excludes other groups",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "vip", constant.ChannelTypeCodex, 10, 100, common.ChannelStatusEnabled, true)
+				insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeOpenAI, 1, 0, common.ChannelStatusEnabled, true)
+			},
+			groups:   []string{"default"},
+			expected: constant.ChannelTypeOpenAI,
+			found:    true,
+		},
+		{
+			name: "disabled candidates are ignored",
+			setup: func(t *testing.T) {
+				insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeCodex, 10, 100, common.ChannelStatusEnabled, false)
+				insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeOpenAI, 1, 0, common.ChannelStatusManuallyDisabled, true)
+			},
+			groups: []string{"default"},
+			found:  false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			clearPreferredOwnerTables(t)
+			tt.setup(t)
+
+			owners, err := GetPreferredModelOwnerChannelTypes([]string{modelName}, tt.groups)
+			require.NoError(t, err)
+
+			got, ok := owners[modelName]
+			require.Equal(t, tt.found, ok)
+			if tt.found {
+				require.Equal(t, tt.expected, got)
+			}
+		})
+	}
+}

model/option.go

@@ -12,6 +12,7 @@ import (
 	"github.com/QuantumNous/new-api/setting/performance_setting"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
 	"github.com/QuantumNous/new-api/setting/system_setting"
+	"gorm.io/gorm"
 )
 
 type Option struct {
@@ -106,18 +107,13 @@ func InitOptionMap() {
 	common.OptionMap["WaffoUnitPrice"] = strconv.FormatFloat(setting.WaffoUnitPrice, 'f', -1, 64)
 	common.OptionMap["WaffoMinTopUp"] = strconv.Itoa(setting.WaffoMinTopUp)
 	common.OptionMap["WaffoPayMethods"] = setting.WaffoPayMethods2JsonString()
-	common.OptionMap["WaffoPancakeEnabled"] = strconv.FormatBool(setting.WaffoPancakeEnabled)
-	common.OptionMap["WaffoPancakeSandbox"] = strconv.FormatBool(setting.WaffoPancakeSandbox)
 	common.OptionMap["WaffoPancakeMerchantID"] = setting.WaffoPancakeMerchantID
 	common.OptionMap["WaffoPancakePrivateKey"] = setting.WaffoPancakePrivateKey
-	common.OptionMap["WaffoPancakeWebhookPublicKey"] = setting.WaffoPancakeWebhookPublicKey
-	common.OptionMap["WaffoPancakeWebhookTestKey"] = setting.WaffoPancakeWebhookTestKey
-	common.OptionMap["WaffoPancakeStoreID"] = setting.WaffoPancakeStoreID
-	common.OptionMap["WaffoPancakeProductID"] = setting.WaffoPancakeProductID
 	common.OptionMap["WaffoPancakeReturnURL"] = setting.WaffoPancakeReturnURL
-	common.OptionMap["WaffoPancakeCurrency"] = setting.WaffoPancakeCurrency
 	common.OptionMap["WaffoPancakeUnitPrice"] = strconv.FormatFloat(setting.WaffoPancakeUnitPrice, 'f', -1, 64)
 	common.OptionMap["WaffoPancakeMinTopUp"] = strconv.Itoa(setting.WaffoPancakeMinTopUp)
+	common.OptionMap["WaffoPancakeStoreID"] = setting.WaffoPancakeStoreID
+	common.OptionMap["WaffoPancakeProductID"] = setting.WaffoPancakeProductID
 	common.OptionMap["TopupGroupRatio"] = common.TopupGroupRatio2JSONString()
 	common.OptionMap["Chats"] = setting.Chats2JsonString()
 	common.OptionMap["AutoGroups"] = setting.AutoGroups2JsonString()
@@ -222,6 +218,39 @@ func UpdateOption(key string, value string) error {
 	return updateOptionMap(key, value)
 }
 
+// UpdateOptionsBulk persists multiple key/value pairs in a single database
+// transaction, then dispatches them through updateOptionMap in one pass. If
+// any DB write fails the whole transaction rolls back and no in-memory state
+// is touched — safe for callers that must commit a set of related options
+// atomically (e.g. payment gateway binding).
+func UpdateOptionsBulk(values map[string]string) error {
+	if len(values) == 0 {
+		return nil
+	}
+	err := DB.Transaction(func(tx *gorm.DB) error {
+		for k, v := range values {
+			option := Option{Key: k}
+			if err := tx.FirstOrCreate(&option, Option{Key: k}).Error; err != nil {
+				return err
+			}
+			option.Value = v
+			if err := tx.Save(&option).Error; err != nil {
+				return err
+			}
+		}
+		return nil
+	})
+	if err != nil {
+		return err
+	}
+	for k, v := range values {
+		if err := updateOptionMap(k, v); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
 func updateOptionMap(key string, value string) (err error) {
 	common.OptionMapRWMutex.Lock()
 	defer common.OptionMapRWMutex.Unlock()
@@ -419,26 +448,16 @@ func updateOptionMap(key string, value string) (err error) {
 		setting.WaffoUnitPrice, _ = strconv.ParseFloat(value, 64)
 	case "WaffoMinTopUp":
 		setting.WaffoMinTopUp, _ = strconv.Atoi(value)
-	case "WaffoPancakeEnabled":
-		setting.WaffoPancakeEnabled = value == "true"
-	case "WaffoPancakeSandbox":
-		setting.WaffoPancakeSandbox = value == "true"
 	case "WaffoPancakeMerchantID":
 		setting.WaffoPancakeMerchantID = value
 	case "WaffoPancakePrivateKey":
 		setting.WaffoPancakePrivateKey = value
-	case "WaffoPancakeWebhookPublicKey":
-		setting.WaffoPancakeWebhookPublicKey = value
-	case "WaffoPancakeWebhookTestKey":
-		setting.WaffoPancakeWebhookTestKey = value
+	case "WaffoPancakeReturnURL":
+		setting.WaffoPancakeReturnURL = value
 	case "WaffoPancakeStoreID":
 		setting.WaffoPancakeStoreID = value
 	case "WaffoPancakeProductID":
 		setting.WaffoPancakeProductID = value
-	case "WaffoPancakeReturnURL":
-		setting.WaffoPancakeReturnURL = value
-	case "WaffoPancakeCurrency":
-		setting.WaffoPancakeCurrency = value
 	case "WaffoPancakeUnitPrice":
 		setting.WaffoPancakeUnitPrice, _ = strconv.ParseFloat(value, 64)
 	case "WaffoPancakeMinTopUp":

model/perf_metric.go

@@ -37,13 +37,13 @@ func UpsertPerfMetric(metric *PerfMetric) error {
 			{Name: "bucket_ts"},
 		},
 		DoUpdates: clause.Assignments(map[string]interface{}{
-			"request_count":    gorm.Expr("request_count + ?", metric.RequestCount),
-			"success_count":    gorm.Expr("success_count + ?", metric.SuccessCount),
-			"total_latency_ms": gorm.Expr("total_latency_ms + ?", metric.TotalLatencyMs),
-			"ttft_sum_ms":      gorm.Expr("ttft_sum_ms + ?", metric.TtftSumMs),
-			"ttft_count":       gorm.Expr("ttft_count + ?", metric.TtftCount),
-			"output_tokens":    gorm.Expr("output_tokens + ?", metric.OutputTokens),
-			"generation_ms":    gorm.Expr("generation_ms + ?", metric.GenerationMs),
+			"request_count":    gorm.Expr("perf_metrics.request_count + ?", metric.RequestCount),
+			"success_count":    gorm.Expr("perf_metrics.success_count + ?", metric.SuccessCount),
+			"total_latency_ms": gorm.Expr("perf_metrics.total_latency_ms + ?", metric.TotalLatencyMs),
+			"ttft_sum_ms":      gorm.Expr("perf_metrics.ttft_sum_ms + ?", metric.TtftSumMs),
+			"ttft_count":       gorm.Expr("perf_metrics.ttft_count + ?", metric.TtftCount),
+			"output_tokens":    gorm.Expr("perf_metrics.output_tokens + ?", metric.OutputTokens),
+			"generation_ms":    gorm.Expr("perf_metrics.generation_ms + ?", metric.GenerationMs),
 		}),
 	}).Create(metric).Error
 }
@@ -68,11 +68,18 @@ type PerfMetricSummary struct {
 	GenerationMs   int64  `json:"generation_ms"`
 }
 
-func GetPerfMetricsSummaryAll(startTs int64, endTs int64) ([]PerfMetricSummary, error) {
+func GetPerfMetricsSummaryAll(startTs int64, endTs int64, groups []string) ([]PerfMetricSummary, error) {
 	var summaries []PerfMetricSummary
-	err := DB.Model(&PerfMetric{}).
+	query := DB.Model(&PerfMetric{}).
 		Select("model_name, SUM(request_count) as request_count, SUM(success_count) as success_count, SUM(total_latency_ms) as total_latency_ms, SUM(output_tokens) as output_tokens, SUM(generation_ms) as generation_ms").
-		Where("bucket_ts >= ? AND bucket_ts <= ?", startTs, endTs).
+		Where("bucket_ts >= ? AND bucket_ts <= ?", startTs, endTs)
+	if groups != nil {
+		if len(groups) == 0 {
+			return summaries, nil
+		}
+		query = query.Where(commonGroupCol+" IN ?", groups)
+	}
+	err := query.
 		Group("model_name").
 		Having("SUM(request_count) > 0").
 		Find(&summaries).Error

model/subscription.go

@@ -11,6 +11,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/pkg/cachex"
 	"github.com/samber/hot"
+	"github.com/shopspring/decimal"
 	"gorm.io/gorm"
 )
 
@@ -159,8 +160,11 @@ type SubscriptionPlan struct {
 	Enabled   bool `json:"enabled" gorm:"default:true"`
 	SortOrder int  `json:"sort_order" gorm:"type:int;default:0"`
 
-	StripePriceId  string `json:"stripe_price_id" gorm:"type:varchar(128);default:''"`
-	CreemProductId string `json:"creem_product_id" gorm:"type:varchar(128);default:''"`
+	AllowBalancePay *bool `json:"allow_balance_pay" gorm:"default:true"`
+
+	StripePriceId         string `json:"stripe_price_id" gorm:"type:varchar(128);default:''"`
+	CreemProductId        string `json:"creem_product_id" gorm:"type:varchar(128);default:''"`
+	WaffoPancakeProductId string `json:"waffo_pancake_product_id" gorm:"type:varchar(128);default:''"`
 
 	// Max purchases per user (0 = unlimited)
 	MaxPurchasePerUser int `json:"max_purchase_per_user" gorm:"type:int;default:0"`
@@ -191,6 +195,12 @@ func (p *SubscriptionPlan) BeforeUpdate(tx *gorm.DB) error {
 	return nil
 }
 
+func (p *SubscriptionPlan) NormalizeDefaults() {
+	if p.AllowBalancePay == nil {
+		p.AllowBalancePay = common.GetPointer(true)
+	}
+}
+
 // Subscription order (payment -> webhook -> create UserSubscription)
 type SubscriptionOrder struct {
 	Id     int     `json:"id"`
@@ -358,6 +368,7 @@ func getSubscriptionPlanByIdTx(tx *gorm.DB, id int) (*SubscriptionPlan, error) {
 	key := subscriptionPlanCacheKey(id)
 	if key != "" {
 		if cached, found, err := getSubscriptionPlanCache().Get(key); err == nil && found {
+			cached.NormalizeDefaults()
 			return &cached, nil
 		}
 	}
@@ -369,6 +380,7 @@ func getSubscriptionPlanByIdTx(tx *gorm.DB, id int) (*SubscriptionPlan, error) {
 	if err := query.Where("id = ?", id).First(&plan).Error; err != nil {
 		return nil, err
 	}
+	plan.NormalizeDefaults()
 	_ = getSubscriptionPlanCache().SetWithTTL(key, plan, subscriptionPlanCacheTTL())
 	return &plan, nil
 }
@@ -664,6 +676,109 @@ func AdminBindSubscription(userId int, planId int, sourceNote string) (string, e
 	return "", nil
 }
 
+func calcSubscriptionBalanceQuota(priceAmount float64) (int, error) {
+	if priceAmount <= 0 {
+		return 0, nil
+	}
+	if common.QuotaPerUnit <= 0 {
+		return 0, errors.New("额度单位配置错误")
+	}
+	quota := decimal.NewFromFloat(priceAmount).
+		Mul(decimal.NewFromFloat(common.QuotaPerUnit)).
+		Ceil().
+		IntPart()
+	return int(quota), nil
+}
+
+// PurchaseSubscriptionWithBalance creates a subscription by deducting the user's wallet quota.
+func PurchaseSubscriptionWithBalance(userId int, planId int) error {
+	if userId <= 0 || planId <= 0 {
+		return errors.New("invalid userId or planId")
+	}
+
+	var logPlanTitle string
+	var logMoney float64
+	var chargedQuota int
+	var upgradeGroup string
+	err := DB.Transaction(func(tx *gorm.DB) error {
+		plan, err := getSubscriptionPlanByIdTx(tx, planId)
+		if err != nil {
+			return err
+		}
+		if !plan.Enabled {
+			return errors.New("套餐未启用")
+		}
+		if plan.PriceAmount < 0 {
+			return errors.New("套餐价格不能为负数")
+		}
+		if plan.AllowBalancePay != nil && !*plan.AllowBalancePay {
+			return errors.New("该套餐不允许使用余额兑换")
+		}
+
+		requiredQuota, err := calcSubscriptionBalanceQuota(plan.PriceAmount)
+		if err != nil {
+			return err
+		}
+
+		var user User
+		if err := tx.Set("gorm:query_option", "FOR UPDATE").Where("id = ?", userId).First(&user).Error; err != nil {
+			return err
+		}
+		if requiredQuota > 0 && user.Quota < requiredQuota {
+			return errors.New("余额不足")
+		}
+		if requiredQuota > 0 {
+			if err := tx.Model(&User{}).Where("id = ?", userId).
+				Update("quota", gorm.Expr("quota - ?", requiredQuota)).Error; err != nil {
+				return err
+			}
+		}
+
+		if _, err := CreateUserSubscriptionFromPlanTx(tx, userId, plan, PaymentMethodBalance); err != nil {
+			return err
+		}
+
+		now := common.GetTimestamp()
+		tradeNo := fmt.Sprintf("SUBBALUSR%dNO%s%d", userId, common.GetRandomString(6), time.Now().UnixNano())
+		order := &SubscriptionOrder{
+			UserId:          userId,
+			PlanId:          plan.Id,
+			Money:           plan.PriceAmount,
+			TradeNo:         tradeNo,
+			PaymentMethod:   PaymentMethodBalance,
+			PaymentProvider: PaymentProviderBalance,
+			Status:          common.TopUpStatusSuccess,
+			CreateTime:      now,
+			CompleteTime:    now,
+			ProviderPayload: fmt.Sprintf("charged_quota=%d", requiredQuota),
+		}
+		if err := tx.Create(order).Error; err != nil {
+			return err
+		}
+
+		logPlanTitle = plan.Title
+		logMoney = plan.PriceAmount
+		chargedQuota = requiredQuota
+		upgradeGroup = strings.TrimSpace(plan.UpgradeGroup)
+		return nil
+	})
+	if err != nil {
+		return err
+	}
+
+	if chargedQuota > 0 {
+		if err := cacheDecrUserQuota(userId, int64(chargedQuota)); err != nil {
+			common.SysLog("failed to decrease user quota cache after subscription balance purchase: " + err.Error())
+		}
+	}
+	if upgradeGroup != "" {
+		_ = UpdateUserGroupCache(userId, upgradeGroup)
+	}
+	msg := fmt.Sprintf("使用余额购买订阅成功，套餐: %s，支付金额: %.2f，扣除额度: %d", logPlanTitle, logMoney, chargedQuota)
+	RecordLog(userId, LogTypeTopup, msg)
+	return nil
+}
+
 // GetAllActiveUserSubscriptions returns all active subscriptions for a user.
 func GetAllActiveUserSubscriptions(userId int) ([]SubscriptionSummary, error) {
 	if userId <= 0 {

model/task_cas_test.go

@@ -26,6 +26,7 @@ func TestMain(m *testing.M) {
 	common.RedisEnabled = false
 	common.BatchUpdateEnabled = false
 	common.LogConsumeEnabled = true
+	initCol()
 
 	sqlDB, err := db.DB()
 	if err != nil {
@@ -39,10 +40,12 @@ func TestMain(m *testing.M) {
 		&Token{},
 		&Log{},
 		&Channel{},
+		&Ability{},
 		&TopUp{},
 		&SubscriptionPlan{},
 		&SubscriptionOrder{},
 		&UserSubscription{},
+		&PerfMetric{},
 	); err != nil {
 		panic("failed to migrate: " + err.Error())
 	}
@@ -58,10 +61,12 @@ func truncateTables(t *testing.T) {
 		DB.Exec("DELETE FROM tokens")
 		DB.Exec("DELETE FROM logs")
 		DB.Exec("DELETE FROM channels")
+		DB.Exec("DELETE FROM abilities")
 		DB.Exec("DELETE FROM top_ups")
 		DB.Exec("DELETE FROM subscription_orders")
 		DB.Exec("DELETE FROM subscription_plans")
 		DB.Exec("DELETE FROM user_subscriptions")
+		DB.Exec("DELETE FROM perf_metrics")
 	})
 }
 

model/topup.go

@@ -29,6 +29,7 @@ const (
 	PaymentMethodCreem        = "creem"
 	PaymentMethodWaffo        = "waffo"
 	PaymentMethodWaffoPancake = "waffo_pancake"
+	PaymentMethodBalance      = "balance"
 )
 
 const (
@@ -37,6 +38,7 @@ const (
 	PaymentProviderCreem        = "creem"
 	PaymentProviderWaffo        = "waffo"
 	PaymentProviderWaffoPancake = "waffo_pancake"
+	PaymentProviderBalance      = "balance"
 )
 
 var (

model/user.go

@@ -35,8 +35,8 @@ type User struct {
 	OidcId           string         `json:"oidc_id" gorm:"column:oidc_id;index"`
 	WeChatId         string         `json:"wechat_id" gorm:"column:wechat_id;index"`
 	TelegramId       string         `json:"telegram_id" gorm:"column:telegram_id;index"`
-	VerificationCode string         `json:"verification_code" gorm:"-:all"`                                    // this field is only for Email verification, don't save it to database!
-	AccessToken      *string        `json:"access_token" gorm:"type:char(32);column:access_token;uniqueIndex"` // this token is for system management
+	VerificationCode string         `json:"verification_code" gorm:"-:all"`                         // this field is only for Email verification, don't save it to database!
+	AccessToken      *string        `json:"-" gorm:"type:char(32);column:access_token;uniqueIndex"` // this token is for system management
 	Quota            int            `json:"quota" gorm:"type:int;default:0"`
 	UsedQuota        int            `json:"used_quota" gorm:"type:int;default:0;column:used_quota"` // used quota
 	RequestCount     int            `json:"request_count" gorm:"type:int;default:0;"`               // request number
@@ -225,7 +225,7 @@ func GetAllUsers(pageInfo *common.PageInfo) (users []*User, total int64, err err
 	return users, total, nil
 }
 
-func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User, int64, error) {
+func SearchUsers(keyword string, group string, role *int, status *int, startIdx int, num int) ([]*User, int64, error) {
 	var users []*User
 	var total int64
 	var err error
@@ -246,28 +246,25 @@ func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User,
 
 	// 构建搜索条件
 	likeCondition := "username LIKE ? OR email LIKE ? OR display_name LIKE ?"
+	likeArgs := []interface{}{"%" + keyword + "%", "%" + keyword + "%", "%" + keyword + "%"}
 
 	// 尝试将关键字转换为整数ID
 	keywordInt, err := strconv.Atoi(keyword)
 	if err == nil {
 		// 如果是数字，同时搜索ID和其他字段
 		likeCondition = "id = ? OR " + likeCondition
-		if group != "" {
-			query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
-				keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
-		} else {
-			query = query.Where(likeCondition,
-				keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
-		}
-	} else {
-		// 非数字关键字，只搜索字符串字段
-		if group != "" {
-			query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
-				"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
-		} else {
-			query = query.Where(likeCondition,
-				"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
-		}
+		likeArgs = append([]interface{}{keywordInt}, likeArgs...)
+	}
+
+	query = query.Where("("+likeCondition+")", likeArgs...)
+	if group != "" {
+		query = query.Where(commonGroupCol+" = ?", group)
+	}
+	if role != nil {
+		query = query.Where("role = ?", *role)
+	}
+	if status != nil {
+		query = query.Where("status = ?", *status)
 	}
 
 	// 获取总数
@@ -987,6 +984,23 @@ func updateUserUsedQuotaAndRequestCount(id int, quota int, count int) {
 	//}
 }
 
+func updateUserQuotaUsedQuotaAndRequestCount(id int, quota int, usedQuota int, requestCount int) {
+	if quota == 0 && usedQuota == 0 && requestCount == 0 {
+		return
+	}
+
+	err := DB.Model(&User{}).Where("id = ?", id).Updates(
+		map[string]interface{}{
+			"quota":         gorm.Expr("quota + ?", quota),
+			"used_quota":    gorm.Expr("used_quota + ?", usedQuota),
+			"request_count": gorm.Expr("request_count + ?", requestCount),
+		},
+	).Error
+	if err != nil {
+		common.SysLog("failed to batch update user quota, used quota and request count: " + err.Error())
+	}
+}
+
 func updateUserUsedQuota(id int, quota int) {
 	err := DB.Model(&User{}).Where("id = ?", id).Updates(
 		map[string]interface{}{

model/utils.go

@@ -67,33 +67,48 @@ func batchUpdate() {
 	}
 
 	common.SysLog("batch update started")
+	stores := make([]map[int]int, BatchUpdateTypeCount)
 	for i := 0; i < BatchUpdateTypeCount; i++ {
 		batchUpdateLocks[i].Lock()
-		store := batchUpdateStores[i]
+		stores[i] = batchUpdateStores[i]
 		batchUpdateStores[i] = make(map[int]int)
 		batchUpdateLocks[i].Unlock()
-		// TODO: maybe we can combine updates with same key?
+	}
+
+	for i, store := range stores {
+		if i == BatchUpdateTypeUserQuota || i == BatchUpdateTypeUsedQuota || i == BatchUpdateTypeRequestCount {
+			continue
+		}
 		for key, value := range store {
 			switch i {
-			case BatchUpdateTypeUserQuota:
-				err := increaseUserQuota(key, value)
-				if err != nil {
-					common.SysLog("failed to batch update user quota: " + err.Error())
-				}
 			case BatchUpdateTypeTokenQuota:
 				err := increaseTokenQuota(key, value)
 				if err != nil {
 					common.SysLog("failed to batch update token quota: " + err.Error())
 				}
-			case BatchUpdateTypeUsedQuota:
-				updateUserUsedQuota(key, value)
-			case BatchUpdateTypeRequestCount:
-				updateUserRequestCount(key, value)
 			case BatchUpdateTypeChannelUsedQuota:
 				updateChannelUsedQuota(key, value)
 			}
 		}
 	}
+
+	userQuotaStore := stores[BatchUpdateTypeUserQuota]
+	usedQuotaStore := stores[BatchUpdateTypeUsedQuota]
+	requestCountStore := stores[BatchUpdateTypeRequestCount]
+
+	userIDs := make(map[int]struct{}, len(userQuotaStore)+len(usedQuotaStore)+len(requestCountStore))
+	for key := range userQuotaStore {
+		userIDs[key] = struct{}{}
+	}
+	for key := range usedQuotaStore {
+		userIDs[key] = struct{}{}
+	}
+	for key := range requestCountStore {
+		userIDs[key] = struct{}{}
+	}
+	for key := range userIDs {
+		updateUserQuotaUsedQuotaAndRequestCount(key, userQuotaStore[key], usedQuotaStore[key], requestCountStore[key])
+	}
 	common.SysLog("batch update finished")
 }
 

pkg/perf_metrics/metrics.go

@@ -122,7 +122,7 @@ func Query(params QueryParams) (QueryResult, error) {
 	return buildQueryResult(params.Model, merged), nil
 }
 
-func QuerySummaryAll(hours int) (SummaryAllResult, error) {
+func QuerySummaryAll(hours int, groups []string) (SummaryAllResult, error) {
 	if hours <= 0 {
 		hours = 24
 	}
@@ -131,8 +131,9 @@ func QuerySummaryAll(hours int) (SummaryAllResult, error) {
 	}
 	endTs := time.Now().Unix()
 	startTs := endTs - int64(hours)*3600
+	allowedGroups := allowedGroupSet(groups)
 
-	rows, err := model.GetPerfMetricsSummaryAll(startTs, endTs)
+	rows, err := model.GetPerfMetricsSummaryAll(startTs, endTs, groups)
 	if err != nil {
 		return SummaryAllResult{}, err
 	}
@@ -153,6 +154,11 @@ func QuerySummaryAll(hours int) (SummaryAllResult, error) {
 		if k.bucketTs < startTs || k.bucketTs > endTs {
 			return true
 		}
+		if allowedGroups != nil {
+			if _, ok := allowedGroups[k.group]; !ok {
+				return true
+			}
+		}
 		snap := value.(*atomicBucket).snapshot()
 		if snap.requestCount == 0 {
 			return true
@@ -193,6 +199,17 @@ func QuerySummaryAll(hours int) (SummaryAllResult, error) {
 	return SummaryAllResult{Models: models}, nil
 }
 
+func allowedGroupSet(groups []string) map[string]struct{} {
+	if groups == nil {
+		return nil
+	}
+	allowed := make(map[string]struct{}, len(groups))
+	for _, group := range groups {
+		allowed[group] = struct{}{}
+	}
+	return allowed
+}
+
 func bucketStart(ts int64) int64 {
 	bucketSeconds := perf_metrics_setting.GetBucketSeconds()
 	if bucketSeconds <= 0 {

relay/channel/ali/image.go

@@ -229,7 +229,7 @@ func asyncTaskWait(c *gin.Context, info *relaycommon.RelayInfo, taskID string) (
 	time.Sleep(time.Duration(5) * time.Second)
 
 	for {
-		logger.LogDebug(c, fmt.Sprintf("asyncTaskWait step %d/%d, wait %d seconds", step, maxStep, waitSeconds))
+		logger.LogDebug(c, "asyncTaskWait step %d/%d, wait %d seconds", step, maxStep, waitSeconds)
 		step++
 		rsp, err, body := updateTask(info, taskID)
 		responseBody = body
@@ -320,11 +320,10 @@ func aliImageHandler(a *Adaptor, c *gin.Context, resp *http.Response, info *rela
 		}
 	}
 
-	//logger.LogDebug(c, "ali_async_task_result: "+string(originRespBody))
 	if a.IsSyncImageModel {
-		logger.LogDebug(c, "ali_sync_image_result: "+string(originRespBody))
+		logger.LogDebug(c, "ali_sync_image_result: %s", originRespBody)
 	} else {
-		logger.LogDebug(c, "ali_async_image_result: "+string(originRespBody))
+		logger.LogDebug(c, "ali_async_image_result: %s", originRespBody)
 	}
 
 	imageResponses := responseAli2OpenAIImage(c, aliResponse, originRespBody, info, responseFormat)

relay/channel/api_request.go

@@ -25,6 +25,23 @@ import (
 	"github.com/gorilla/websocket"
 )
 
+// applyUpstreamContentLength populates req.ContentLength when the upstream
+// body is wrapped in a BodyStorage (see relay/common/outbound_body.go).
+//
+// net/http.NewRequest only auto-detects ContentLength for *bytes.Reader,
+// *bytes.Buffer and *strings.Reader. When the body is a type-erased io.Reader
+// (which is the case for ReaderOnly(BodyStorage)), the Content-Length header
+// would otherwise be omitted, forcing chunked transfer encoding and breaking
+// some upstreams that require an explicit Content-Length.
+func applyUpstreamContentLength(req *http.Request, info *common.RelayInfo) {
+	if info == nil {
+		return
+	}
+	if info.UpstreamRequestBodySize > 0 && req.ContentLength <= 0 {
+		req.ContentLength = info.UpstreamRequestBodySize
+	}
+}
+
 func SetupApiRequestHeader(info *common.RelayInfo, c *gin.Context, req *http.Header) {
 	if info.RelayMode == constant.RelayModeAudioTranscription || info.RelayMode == constant.RelayModeAudioTranslation {
 		// multipart/form-data
@@ -292,13 +309,12 @@ func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody
 	if err != nil {
 		return nil, fmt.Errorf("get request url failed: %w", err)
 	}
-	if common2.DebugEnabled {
-		println("fullRequestURL:", fullRequestURL)
-	}
+	logger.LogDebug(c, "fullRequestURL: %s", fullRequestURL)
 	req, err := http.NewRequest(c.Request.Method, fullRequestURL, requestBody)
 	if err != nil {
 		return nil, fmt.Errorf("new request failed: %w", err)
 	}
+	applyUpstreamContentLength(req, info)
 	headers := req.Header
 	err = a.SetupRequestHeader(c, &headers, info)
 	if err != nil {
@@ -323,13 +339,12 @@ func DoFormRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBod
 	if err != nil {
 		return nil, fmt.Errorf("get request url failed: %w", err)
 	}
-	if common2.DebugEnabled {
-		println("fullRequestURL:", fullRequestURL)
-	}
+	logger.LogDebug(c, "fullRequestURL: %s", fullRequestURL)
 	req, err := http.NewRequest(c.Request.Method, fullRequestURL, requestBody)
 	if err != nil {
 		return nil, fmt.Errorf("new request failed: %w", err)
 	}
+	applyUpstreamContentLength(req, info)
 	// set form data
 	req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
 	headers := req.Header
@@ -388,13 +403,9 @@ func startPingKeepAlive(c *gin.Context, pingInterval time.Duration) context.Canc
 		defer func() {
 			// 增加panic恢复处理
 			if r := recover(); r != nil {
-				if common2.DebugEnabled {
-					println("SSE ping goroutine panic recovered:", fmt.Sprintf("%v", r))
-				}
-			}
-			if common2.DebugEnabled {
-				println("SSE ping goroutine stopped.")
+				logger.LogDebug(c, "SSE ping goroutine panic recovered: %v", r)
 			}
+			logger.LogDebug(c, "SSE ping goroutine stopped")
 		}()
 
 		if pingInterval <= 0 {
@@ -405,15 +416,11 @@ func startPingKeepAlive(c *gin.Context, pingInterval time.Duration) context.Canc
 		// 确保在任何情况下都清理ticker
 		defer func() {
 			ticker.Stop()
-			if common2.DebugEnabled {
-				println("SSE ping ticker stopped")
-			}
+			logger.LogDebug(c, "SSE ping ticker stopped")
 		}()
 
 		var pingMutex sync.Mutex
-		if common2.DebugEnabled {
-			println("SSE ping goroutine started")
-		}
+		logger.LogDebug(c, "SSE ping goroutine started")
 
 		// 增加超时控制，防止goroutine长时间运行
 		maxPingDuration := 120 * time.Minute // 最大ping持续时间
@@ -425,9 +432,7 @@ func startPingKeepAlive(c *gin.Context, pingInterval time.Duration) context.Canc
 			// 发送 ping 数据
 			case <-ticker.C:
 				if err := sendPingData(c, &pingMutex); err != nil {
-					if common2.DebugEnabled {
-						println("SSE ping error, stopping goroutine:", err.Error())
-					}
+					logger.LogDebug(c, "SSE ping error, stopping goroutine: %s", err.Error())
 					return
 				}
 			// 收到退出信号
@@ -438,9 +443,7 @@ func startPingKeepAlive(c *gin.Context, pingInterval time.Duration) context.Canc
 				return
 			// 超时保护，防止goroutine无限运行
 			case <-pingTimeout.C:
-				if common2.DebugEnabled {
-					println("SSE ping goroutine timeout, stopping")
-				}
+				logger.LogDebug(c, "SSE ping goroutine timeout, stopping")
 				return
 			}
 		}
@@ -463,9 +466,7 @@ func sendPingData(c *gin.Context, mutex *sync.Mutex) error {
 			return
 		}
 
-		if common2.DebugEnabled {
-			println("SSE ping data sent.")
-		}
+		logger.LogDebug(c, "SSE ping data sent")
 		done <- nil
 	}()
 
@@ -507,9 +508,7 @@ func doRequest(c *gin.Context, req *http.Request, info *common.RelayInfo) (*http
 			defer func() {
 				if stopPinger != nil {
 					stopPinger()
-					if common2.DebugEnabled {
-						println("SSE ping goroutine stopped by defer")
-					}
+					logger.LogDebug(c, "SSE ping goroutine stopped by defer")
 				}
 			}()
 		}
@@ -542,6 +541,7 @@ func DoTaskApiRequest(a TaskAdaptor, c *gin.Context, info *common.RelayInfo, req
 	if err != nil {
 		return nil, fmt.Errorf("new request failed: %w", err)
 	}
+	applyUpstreamContentLength(req, info)
 	req.GetBody = func() (io.ReadCloser, error) {
 		return io.NopCloser(requestBody), nil
 	}

relay/channel/aws/constants.go

@@ -19,6 +19,7 @@ var awsModelIDMap = map[string]string{
 	"claude-opus-4-5-20251101":   "anthropic.claude-opus-4-5-20251101-v1:0",
 	"claude-opus-4-6":            "anthropic.claude-opus-4-6-v1",
 	"claude-opus-4-7":            "anthropic.claude-opus-4-7",
+	"claude-opus-4-8":            "anthropic.claude-opus-4-8",
 	// Nova models
 	"nova-micro-v1:0":   "amazon.nova-micro-v1:0",
 	"nova-lite-v1:0":    "amazon.nova-lite-v1:0",
@@ -97,6 +98,11 @@ var awsModelCanCrossRegionMap = map[string]map[string]bool{
 		"ap": true,
 		"eu": true,
 	},
+	"anthropic.claude-opus-4-8": {
+		"us": true,
+		"ap": true,
+		"eu": true,
+	},
 	"anthropic.claude-haiku-4-5-20251001-v1:0": {
 		"us": true,
 		"ap": true,

relay/channel/claude/constants.go

@@ -33,6 +33,13 @@ var ModelList = []string{
 	"claude-opus-4-7-medium",
 	"claude-opus-4-7-low",
 	"claude-opus-4-7-thinking",
+	"claude-opus-4-8",
+	"claude-opus-4-8-max",
+	"claude-opus-4-8-xhigh",
+	"claude-opus-4-8-high",
+	"claude-opus-4-8-medium",
+	"claude-opus-4-8-low",
+	"claude-opus-4-8-thinking",
 }
 
 var ChannelName = "claude"

relay/channel/claude/relay-claude.go

@@ -154,14 +154,17 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 	}
 
 	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(textRequest.Model); ok && effortLevel != "" &&
-		(strings.HasPrefix(textRequest.Model, "claude-opus-4-6") || strings.HasPrefix(textRequest.Model, "claude-opus-4-7")) {
+		(strings.HasPrefix(textRequest.Model, "claude-opus-4-6") ||
+			strings.HasPrefix(textRequest.Model, "claude-opus-4-7") ||
+			strings.HasPrefix(textRequest.Model, "claude-opus-4-8")) {
 		claudeRequest.Model = baseModel
 		claudeRequest.Thinking = &dto.Thinking{
 			Type: "adaptive",
 		}
 		claudeRequest.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
-		if strings.HasPrefix(baseModel, "claude-opus-4-7") {
-			// Opus 4.7 rejects non-default temperature/top_p/top_k with 400
+		if strings.HasPrefix(baseModel, "claude-opus-4-7") ||
+			strings.HasPrefix(baseModel, "claude-opus-4-8") {
+			// Opus 4.7/4.8 reject non-default temperature/top_p/top_k with 400
 			// and defaults display to "omitted"; restore the 4.6 visible summary.
 			claudeRequest.Thinking.Display = "summarized"
 			claudeRequest.Temperature = nil
@@ -175,8 +178,9 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 		strings.HasSuffix(textRequest.Model, "-thinking") {
 
 		trimmedModel := strings.TrimSuffix(textRequest.Model, "-thinking")
-		if strings.HasPrefix(trimmedModel, "claude-opus-4-7") {
-			// Opus 4.7 rejects thinking.type="enabled"; use adaptive at high effort.
+		if strings.HasPrefix(trimmedModel, "claude-opus-4-7") ||
+			strings.HasPrefix(trimmedModel, "claude-opus-4-8") {
+			// Opus 4.7/4.8 reject thinking.type="enabled"; use adaptive at high effort.
 			claudeRequest.Thinking = &dto.Thinking{Type: "adaptive", Display: "summarized"}
 			claudeRequest.OutputConfig = json.RawMessage(`{"effort":"high"}`)
 			claudeRequest.Temperature = nil
@@ -442,10 +446,7 @@ func StreamResponseClaude2OpenAI(claudeResponse *dto.ClaudeResponse) *dto.ChatCo
 	tools := make([]dto.ToolCallResponse, 0)
 	fcIdx := 0
 	if claudeResponse.Index != nil {
-		fcIdx = *claudeResponse.Index - 1
-		if fcIdx < 0 {
-			fcIdx = 0
-		}
+		fcIdx = *claudeResponse.Index
 	}
 	var choice dto.ChatCompletionsStreamResponseChoice
 	if claudeResponse.Type == "message_start" {
@@ -949,9 +950,7 @@ func ClaudeHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayI
 	if err != nil {
 		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
 	}
-	if common.DebugEnabled {
-		println("responseBody: ", string(responseBody))
-	}
+	logger.LogDebug(c, "responseBody: %s", responseBody)
 	handleErr := HandleClaudeResponseData(c, info, claudeInfo, resp, responseBody)
 	if handleErr != nil {
 		return nil, handleErr

relay/channel/claude/relay_claude_test.go

@@ -9,6 +9,10 @@ import (
 	"github.com/stretchr/testify/require"
 )
 
+func commonPointer[T any](value T) *T {
+	return &value
+}
+
 func TestFormatClaudeResponseInfo_MessageStart(t *testing.T) {
 	claudeInfo := &ClaudeResponseInfo{
 		Usage: &dto.Usage{},
@@ -310,6 +314,58 @@ func TestRequestOpenAI2ClaudeMessage_IgnoresUnsupportedFileContent(t *testing.T)
 	require.Equal(t, "see attachment", *content[0].Text)
 }
 
+func TestRequestOpenAI2ClaudeMessage_ClaudeOpus48HighUsesAdaptiveThinking(t *testing.T) {
+	request := dto.GeneralOpenAIRequest{
+		Model:       "claude-opus-4-8-high",
+		Temperature: commonPointer(0.7),
+		TopP:        commonPointer(0.9),
+		TopK:        commonPointer(40),
+		Messages: []dto.Message{
+			{
+				Role:    "user",
+				Content: "hello",
+			},
+		},
+	}
+
+	claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
+	require.NoError(t, err)
+	require.Equal(t, "claude-opus-4-8", claudeRequest.Model)
+	require.NotNil(t, claudeRequest.Thinking)
+	require.Equal(t, "adaptive", claudeRequest.Thinking.Type)
+	require.Equal(t, "summarized", claudeRequest.Thinking.Display)
+	require.JSONEq(t, `{"effort":"high"}`, string(claudeRequest.OutputConfig))
+	require.Nil(t, claudeRequest.Temperature)
+	require.Nil(t, claudeRequest.TopP)
+	require.Nil(t, claudeRequest.TopK)
+}
+
+func TestRequestOpenAI2ClaudeMessage_ClaudeOpus48ThinkingUsesAdaptiveHighEffort(t *testing.T) {
+	request := dto.GeneralOpenAIRequest{
+		Model:       "claude-opus-4-8-thinking",
+		Temperature: commonPointer(0.7),
+		TopP:        commonPointer(0.9),
+		TopK:        commonPointer(40),
+		Messages: []dto.Message{
+			{
+				Role:    "user",
+				Content: "hello",
+			},
+		},
+	}
+
+	claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
+	require.NoError(t, err)
+	require.Equal(t, "claude-opus-4-8", claudeRequest.Model)
+	require.NotNil(t, claudeRequest.Thinking)
+	require.Equal(t, "adaptive", claudeRequest.Thinking.Type)
+	require.Equal(t, "summarized", claudeRequest.Thinking.Display)
+	require.JSONEq(t, `{"effort":"high"}`, string(claudeRequest.OutputConfig))
+	require.Nil(t, claudeRequest.Temperature)
+	require.Nil(t, claudeRequest.TopP)
+	require.Nil(t, claudeRequest.TopK)
+}
+
 func TestRequestOpenAI2ClaudeMessage_SupportsPDFFileContent(t *testing.T) {
 	request := dto.GeneralOpenAIRequest{
 		Model: "claude-3-5-sonnet",

relay/channel/cloudflare/relay_cloudflare.go

@@ -30,7 +30,7 @@ func convertCf2CompletionsRequest(textRequest dto.GeneralOpenAIRequest) *CfReque
 }
 
 func cfStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*types.NewAPIError, *dto.Usage) {
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 
 	helper.SetEventStreamHeaders(c)

relay/channel/cohere/relay-cohere.go

@@ -1,7 +1,6 @@
 package cohere
 
 import (
-	"bufio"
 	"encoding/json"
 	"io"
 	"net/http"
@@ -86,7 +85,7 @@ func cohereStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 	createdTime := common.GetTimestamp()
 	usage := &dto.Usage{}
 	responseText := ""
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		if atEOF && len(data) == 0 {
 			return 0, nil, nil
@@ -106,6 +105,9 @@ func cohereStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 			data := scanner.Text()
 			dataChan <- data
 		}
+		if err := scanner.Err(); err != nil {
+			common.SysLog("error reading stream: " + err.Error())
+		}
 		stopChan <- true
 	}()
 	helper.SetEventStreamHeaders(c)

relay/channel/coze/relay-coze.go

@@ -98,7 +98,7 @@ func cozeChatHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Res
 }
 
 func cozeChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 	helper.SetEventStreamHeaders(c)
 	id := helper.GetResponseID(c)

relay/channel/dify/relay-dify.go

@@ -159,9 +159,14 @@ func requestOpenAI2Dify(c *gin.Context, info *relaycommon.RelayInfo, request dto
 					media := mediaContent.GetImageMedia()
 					var file *DifyFile
 					if media.IsRemoteImage() {
-						file.Type = media.MimeType
-						file.TransferMode = "remote_url"
-						file.URL = media.Url
+						// 修复 #2083: 远程图片分支此前未初始化 file，
+						// 导致 file.Type = ... 触发 nil pointer dereference
+						// 而 panic（500: "invalid memory address or nil pointer dereference"）。
+						file = &DifyFile{
+							Type:         media.MimeType,
+							TransferMode: "remote_url",
+							URL:          media.Url,
+						}
 					} else {
 						file = uploadDifyFile(c, info, difyReq.User, mediaContent)
 					}

relay/channel/gemini/relay-gemini-native.go

@@ -26,9 +26,7 @@ func GeminiTextGenerationHandler(c *gin.Context, info *relaycommon.RelayInfo, re
 		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
 	}
 
-	if common.DebugEnabled {
-		println(string(responseBody))
-	}
+	logger.LogDebug(c, "Gemini native response body: %s", responseBody)
 
 	// 解析为 Gemini 原生响应格式
 	var geminiResponse dto.GeminiChatResponse
@@ -57,9 +55,7 @@ func NativeGeminiEmbeddingHandler(c *gin.Context, resp *http.Response, info *rel
 		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
 	}
 
-	if common.DebugEnabled {
-		println(string(responseBody))
-	}
+	logger.LogDebug(c, "Gemini native embedding response body: %s", responseBody)
 
 	usage := service.ResponseText2Usage(c, "", info.UpstreamModelName, info.GetEstimatePromptTokens())
 

relay/channel/gemini/relay-gemini.go

@@ -1079,17 +1079,47 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
 			FinishReason: constant.FinishReasonStop,
 		}
 		if len(candidate.Content.Parts) > 0 {
-			var texts []string
+			// 使用 strings.Builder 直接累积最终 content，避免:
+			//   1) 每张 inline image 生成一次中间 "![image](...)" 字符串
+			//   2) 末尾 strings.Join 再分配一份等大缓冲
+			// Gemini 图片返回时 InlineData.Data 可能是数 MB 的 base64，
+			// 上述两份临时分配在高并发下会显著放大堆驻留。
+			var content strings.Builder
+			var inlineGrow int
+			for _, part := range candidate.Content.Parts {
+				if part.InlineData != nil {
+					inlineGrow += len(part.InlineData.MimeType) + len(part.InlineData.Data) + 32
+				}
+			}
+			if inlineGrow > 0 {
+				content.Grow(inlineGrow)
+			}
+			appended := 0
+			writeSep := func() {
+				if appended > 0 {
+					content.WriteByte('\n')
+				}
+				appended++
+			}
 			var toolCalls []dto.ToolCallResponse
 			for _, part := range candidate.Content.Parts {
 				if part.InlineData != nil {
 					// 媒体内容
 					if strings.HasPrefix(part.InlineData.MimeType, "image") {
-						imgText := "![image](data:" + part.InlineData.MimeType + ";base64," + part.InlineData.Data + ")"
-						texts = append(texts, imgText)
+						writeSep()
+						content.WriteString("![image](data:")
+						content.WriteString(part.InlineData.MimeType)
+						content.WriteString(";base64,")
+						content.WriteString(part.InlineData.Data)
+						content.WriteByte(')')
 					} else {
 						// 其他媒体类型，直接显示链接
-						texts = append(texts, fmt.Sprintf("[media](data:%s;base64,%s)", part.InlineData.MimeType, part.InlineData.Data))
+						writeSep()
+						content.WriteString("[media](data:")
+						content.WriteString(part.InlineData.MimeType)
+						content.WriteString(";base64,")
+						content.WriteString(part.InlineData.Data)
+						content.WriteByte(')')
 					}
 				} else if part.FunctionCall != nil {
 					choice.FinishReason = constant.FinishReasonToolCalls
@@ -1100,13 +1130,22 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
 					choice.Message.ReasoningContent = &part.Text
 				} else {
 					if part.ExecutableCode != nil {
-						texts = append(texts, "```"+part.ExecutableCode.Language+"\n"+part.ExecutableCode.Code+"\n```")
+						writeSep()
+						content.WriteString("```")
+						content.WriteString(part.ExecutableCode.Language)
+						content.WriteByte('\n')
+						content.WriteString(part.ExecutableCode.Code)
+						content.WriteString("\n```")
 					} else if part.CodeExecutionResult != nil {
-						texts = append(texts, "```output\n"+part.CodeExecutionResult.Output+"\n```")
+						writeSep()
+						content.WriteString("```output\n")
+						content.WriteString(part.CodeExecutionResult.Output)
+						content.WriteString("\n```")
 					} else {
 						// 过滤掉空行
 						if part.Text != "\n" {
-							texts = append(texts, part.Text)
+							writeSep()
+							content.WriteString(part.Text)
 						}
 					}
 				}
@@ -1115,7 +1154,7 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
 				choice.Message.SetToolCalls(toolCalls)
 				isToolCall = true
 			}
-			choice.Message.SetStringContent(strings.Join(texts, "\n"))
+			choice.Message.SetStringContent(content.String())
 
 		}
 		if candidate.FinishReason != nil {
@@ -1169,7 +1208,25 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
 				//Role: "assistant",
 			},
 		}
-		var texts []string
+		// 使用 strings.Builder 直接累积 delta content，避免每张 image / 每个
+		// 文本片段都先 `+` 拼出一份临时 string，再 strings.Join 再拷贝一遍。
+		var content strings.Builder
+		var inlineGrow int
+		for _, part := range candidate.Content.Parts {
+			if part.InlineData != nil {
+				inlineGrow += len(part.InlineData.MimeType) + len(part.InlineData.Data) + 32
+			}
+		}
+		if inlineGrow > 0 {
+			content.Grow(inlineGrow)
+		}
+		appended := 0
+		writeSep := func() {
+			if appended > 0 {
+				content.WriteByte('\n')
+			}
+			appended++
+		}
 		isTools := false
 		isThought := false
 		if candidate.FinishReason != nil {
@@ -1207,8 +1264,12 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
 		for _, part := range candidate.Content.Parts {
 			if part.InlineData != nil {
 				if strings.HasPrefix(part.InlineData.MimeType, "image") {
-					imgText := "![image](data:" + part.InlineData.MimeType + ";base64," + part.InlineData.Data + ")"
-					texts = append(texts, imgText)
+					writeSep()
+					content.WriteString("![image](data:")
+					content.WriteString(part.InlineData.MimeType)
+					content.WriteString(";base64,")
+					content.WriteString(part.InlineData.Data)
+					content.WriteByte(')')
 				}
 			} else if part.FunctionCall != nil {
 				isTools = true
@@ -1219,23 +1280,33 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
 
 			} else if part.Thought {
 				isThought = true
-				texts = append(texts, part.Text)
+				writeSep()
+				content.WriteString(part.Text)
 			} else {
 				if part.ExecutableCode != nil {
-					texts = append(texts, "```"+part.ExecutableCode.Language+"\n"+part.ExecutableCode.Code+"\n```\n")
+					writeSep()
+					content.WriteString("```")
+					content.WriteString(part.ExecutableCode.Language)
+					content.WriteByte('\n')
+					content.WriteString(part.ExecutableCode.Code)
+					content.WriteString("\n```\n")
 				} else if part.CodeExecutionResult != nil {
-					texts = append(texts, "```output\n"+part.CodeExecutionResult.Output+"\n```\n")
+					writeSep()
+					content.WriteString("```output\n")
+					content.WriteString(part.CodeExecutionResult.Output)
+					content.WriteString("\n```\n")
 				} else {
 					if part.Text != "\n" {
-						texts = append(texts, part.Text)
+						writeSep()
+						content.WriteString(part.Text)
 					}
 				}
 			}
 		}
 		if isThought {
-			choice.Delta.SetReasoningContent(strings.Join(texts, "\n"))
+			choice.Delta.SetReasoningContent(content.String())
 		} else {
-			choice.Delta.SetContentString(strings.Join(texts, "\n"))
+			choice.Delta.SetContentString(content.String())
 		}
 		if isTools {
 			choice.FinishReason = &constant.FinishReasonToolCalls
@@ -1339,6 +1410,14 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 		response.Id = id
 		response.Created = createAt
 		response.Model = info.UpstreamModelName
+		if response.IsToolCall() {
+			finishReason = constant.FinishReasonToolCalls
+			if info.RelayFormat == types.RelayFormatClaude {
+				for choiceIdx := range response.Choices {
+					response.Choices[choiceIdx].FinishReason = nil
+				}
+			}
+		}
 		for choiceIdx := range response.Choices {
 			choiceKey := response.Choices[choiceIdx].Index
 			for toolIdx := range response.Choices[choiceIdx].Delta.ToolCalls {
@@ -1362,7 +1441,7 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 			}
 		}
 
-		logger.LogDebug(c, fmt.Sprintf("info.SendResponseCount = %d", info.SendResponseCount))
+		logger.LogDebug(c, "info.SendResponseCount = %d", info.SendResponseCount)
 		if info.SendResponseCount == 0 {
 			// send first response
 			emptyResponse := helper.GenerateStartEmptyResponse(id, createAt, info.UpstreamModelName, nil)
@@ -1399,7 +1478,9 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 			logger.LogError(c, err.Error())
 		}
 		if isStop {
-			_ = handleStream(c, info, helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason))
+			if info.RelayFormat != types.RelayFormatClaude {
+				_ = handleStream(c, info, helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason))
+			}
 		}
 		return true
 	})
@@ -1409,6 +1490,10 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
 	}
 
 	response := helper.GenerateFinalUsageResponse(id, createAt, info.UpstreamModelName, *usage)
+	if info.RelayFormat == types.RelayFormatClaude && info.ClaudeConvertInfo != nil && !info.ClaudeConvertInfo.Done {
+		response = helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason)
+		response.Usage = usage
+	}
 	handleErr := handleFinalStream(c, info, response)
 	if handleErr != nil {
 		common.SysLog("send final response failed: " + handleErr.Error())
@@ -1422,9 +1507,7 @@ func GeminiChatHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.R
 		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
 	}
 	service.CloseResponseBodyGracefully(resp)
-	if common.DebugEnabled {
-		println(string(responseBody))
-	}
+	logger.LogDebug(c, "Gemini response body: %s", responseBody)
 	var geminiResponse dto.GeminiChatResponse
 	err = common.Unmarshal(responseBody, &geminiResponse)
 	if err != nil {

relay/channel/moonshot/adaptor.go

@@ -5,7 +5,9 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"strings"
 
+	"github.com/QuantumNous/new-api/common"
 	channelconstant "github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/relay/channel"
@@ -79,9 +81,23 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
 }
 
 func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error) {
+	if request.Temperature != nil && isTemperatureOneOnlyModel(getUpstreamModelName(info, request.Model)) && *request.Temperature != 1.0 {
+		request.Temperature = common.GetPointer[float64](1.0)
+	}
 	return request, nil
 }
 
+func getUpstreamModelName(info *relaycommon.RelayInfo, fallback string) string {
+	if info != nil && info.ChannelMeta != nil && info.UpstreamModelName != "" {
+		return info.UpstreamModelName
+	}
+	return fallback
+}
+
+func isTemperatureOneOnlyModel(model string) bool {
+	return strings.EqualFold(model, "kimi-k2.6")
+}
+
 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
 	// TODO implement me
 	return nil, errors.New("not implemented")

relay/channel/moonshot/adaptor_test.go

@@ -0,0 +1,68 @@
+package moonshot
+
+import (
+	"testing"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/dto"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/stretchr/testify/require"
+)
+
+func TestConvertOpenAIRequestKimiK26UsesOnlyAllowedTemperature(t *testing.T) {
+	request := &dto.GeneralOpenAIRequest{
+		Model:       "kimi-k2.6",
+		Temperature: common.GetPointer[float64](0.7),
+	}
+	info := &relaycommon.RelayInfo{
+		ChannelMeta: &relaycommon.ChannelMeta{
+			UpstreamModelName: "kimi-k2.6",
+		},
+	}
+
+	converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
+
+	require.NoError(t, err)
+	convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
+	require.True(t, ok)
+	require.NotNil(t, convertedRequest.Temperature)
+	require.Equal(t, 1.0, *convertedRequest.Temperature)
+}
+
+func TestConvertOpenAIRequestKimiK26KeepsOmittedTemperatureOmitted(t *testing.T) {
+	request := &dto.GeneralOpenAIRequest{
+		Model: "kimi-k2.6",
+	}
+	info := &relaycommon.RelayInfo{
+		ChannelMeta: &relaycommon.ChannelMeta{
+			UpstreamModelName: "kimi-k2.6",
+		},
+	}
+
+	converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
+
+	require.NoError(t, err)
+	convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
+	require.True(t, ok)
+	require.Nil(t, convertedRequest.Temperature)
+}
+
+func TestConvertOpenAIRequestOtherMoonshotModelKeepsTemperature(t *testing.T) {
+	request := &dto.GeneralOpenAIRequest{
+		Model:       "kimi-k2.5",
+		Temperature: common.GetPointer[float64](0.7),
+	}
+	info := &relaycommon.RelayInfo{
+		ChannelMeta: &relaycommon.ChannelMeta{
+			UpstreamModelName: "kimi-k2.5",
+		},
+	}
+
+	converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
+
+	require.NoError(t, err)
+	convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
+	require.True(t, ok)
+	require.NotNil(t, convertedRequest.Temperature)
+	require.Equal(t, 0.7, *convertedRequest.Temperature)
+}

relay/channel/ollama/relay-ollama.go

@@ -1,7 +1,6 @@
 package ollama
 
 import (
-	"bufio"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -12,6 +11,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/dto"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/types"
 
@@ -397,7 +397,7 @@ func PullOllamaModelStream(baseURL, apiKey, modelName string, progressCallback f
 	}
 
 	// 读取流式响应
-	scanner := bufio.NewScanner(response.Body)
+	scanner := helper.NewStreamScanner(response.Body)
 	successful := false
 	for scanner.Scan() {
 		line := scanner.Text()

relay/channel/ollama/stream.go

@@ -1,7 +1,6 @@
 package ollama
 
 import (
-	"bufio"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -70,7 +69,7 @@ func ollamaStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 	defer service.CloseResponseBodyGracefully(resp)
 
 	helper.SetEventStreamHeaders(c)
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	usage := &dto.Usage{}
 	var model = info.UpstreamModelName
 	var responseId = common.GetUUID()

relay/channel/openai/adaptor.go

@@ -9,6 +9,7 @@ import (
 	"mime/multipart"
 	"net/http"
 	"net/textproto"
+	"net/url"
 	"path/filepath"
 	"strings"
 
@@ -310,18 +311,20 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 		}
 
 	}
-	if strings.HasPrefix(info.UpstreamModelName, "o") || strings.HasPrefix(info.UpstreamModelName, "gpt-5") {
+	isOModel := dto.IsOpenAIReasoningOModel(info.UpstreamModelName)
+	isGPT5Model := dto.IsOpenAIGPT5Model(info.UpstreamModelName)
+	if isOModel || isGPT5Model {
 		if lo.FromPtrOr(request.MaxCompletionTokens, uint(0)) == 0 && lo.FromPtrOr(request.MaxTokens, uint(0)) != 0 {
 			request.MaxCompletionTokens = request.MaxTokens
 			request.MaxTokens = nil
 		}
 
-		if strings.HasPrefix(info.UpstreamModelName, "o") {
+		if isOModel {
 			request.Temperature = nil
 		}
 
 		// gpt-5系列模型适配 归零不再支持的参数
-		if strings.HasPrefix(info.UpstreamModelName, "gpt-5") {
+		if isGPT5Model {
 			request.Temperature = nil
 			request.TopP = nil
 			request.LogProbs = nil
@@ -377,7 +380,7 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 		}
 
 		// 打印类似 curl 命令格式的信息
-		logger.LogDebug(c.Request.Context(), fmt.Sprintf("--form 'model=\"%s\"'", request.Model))
+		logger.LogDebug(c.Request.Context(), "--form 'model=\"%s\"'", request.Model)
 
 		// 遍历表单字段并打印输出
 		for key, values := range formData.Value {
@@ -386,7 +389,7 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 			}
 			for _, value := range values {
 				writer.WriteField(key, value)
-				logger.LogDebug(c.Request.Context(), fmt.Sprintf("--form '%s=\"%s\"'", key, value))
+				logger.LogDebug(c.Request.Context(), "--form '%s=\"%s\"'", key, value)
 			}
 		}
 
@@ -398,8 +401,8 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 
 		// 使用 formData 中的第一个文件
 		fileHeader := fileHeaders[0]
-		logger.LogDebug(c.Request.Context(), fmt.Sprintf("--form 'file=@\"%s\"' (size: %d bytes, content-type: %s)",
-			fileHeader.Filename, fileHeader.Size, fileHeader.Header.Get("Content-Type")))
+		logger.LogDebug(c.Request.Context(), "--form 'file=@\"%s\"' (size: %d bytes, content-type: %s)",
+			fileHeader.Filename, fileHeader.Size, fileHeader.Header.Get("Content-Type"))
 
 		file, err := fileHeader.Open()
 		if err != nil {
@@ -418,7 +421,7 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 		// 关闭 multipart 编写器以设置分界线
 		writer.Close()
 		c.Request.Header.Set("Content-Type", writer.FormDataContentType())
-		logger.LogDebug(c.Request.Context(), fmt.Sprintf("--header 'Content-Type: %s'", writer.FormDataContentType()))
+		logger.LogDebug(c.Request.Context(), "--header 'Content-Type: %s'", writer.FormDataContentType())
 		return &requestBody, nil
 	}
 }
@@ -437,10 +440,13 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 		// 使用已解析的 multipart 表单，避免重复解析
 		mf := c.Request.MultipartForm
 		if mf == nil {
-			if _, err := c.MultipartForm(); err != nil {
-				return nil, errors.New("failed to parse multipart form")
+			form, err := common.ParseMultipartFormReusable(c)
+			if err != nil {
+				return nil, fmt.Errorf("failed to parse multipart form: %w", err)
 			}
-			mf = c.Request.MultipartForm
+			c.Request.MultipartForm = form
+			c.Request.PostForm = url.Values(form.Value)
+			mf = form
 		}
 
 		// 写入所有非文件字段
@@ -623,7 +629,11 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	case relayconstant.RelayModeAudioTranscription:
 		err, usage = OpenaiSTTHandler(c, resp, info, a.ResponseFormat)
 	case relayconstant.RelayModeImagesGenerations, relayconstant.RelayModeImagesEdits:
-		usage, err = OpenaiHandlerWithUsage(c, info, resp)
+		if info.IsStream {
+			usage, err = OpenaiImageStreamHandler(c, info, resp)
+		} else {
+			usage, err = OpenaiImageHandler(c, info, resp)
+		}
 	case relayconstant.RelayModeRerank:
 		usage, err = common_handler.RerankHandler(c, info, resp)
 	case relayconstant.RelayModeResponses:

relay/channel/openai/helper.go

@@ -1,7 +1,6 @@
 package openai
 
 import (
-	"encoding/json"
 	"strings"
 
 	"github.com/QuantumNous/new-api/common"
@@ -92,78 +91,28 @@ func ProcessStreamResponse(streamResponse dto.ChatCompletionsStreamResponse, res
 	return nil
 }
 
-func processTokens(relayMode int, streamItems []string, responseTextBuilder *strings.Builder, toolCount *int) error {
-	streamResp := "[" + strings.Join(streamItems, ",") + "]"
-
+func processTokenData(relayMode int, data string, responseTextBuilder *strings.Builder, toolCount *int) error {
 	switch relayMode {
 	case relayconstant.RelayModeChatCompletions:
-		return processChatCompletions(streamResp, streamItems, responseTextBuilder, toolCount)
+		var streamResponse dto.ChatCompletionsStreamResponse
+		if err := common.UnmarshalJsonStr(data, &streamResponse); err != nil {
+			return err
+		}
+		return ProcessStreamResponse(streamResponse, responseTextBuilder, toolCount)
 	case relayconstant.RelayModeCompletions:
-		return processCompletions(streamResp, streamItems, responseTextBuilder)
+		var streamResponse dto.CompletionsStreamResponse
+		if err := common.UnmarshalJsonStr(data, &streamResponse); err != nil {
+			return err
+		}
+		processCompletionsStreamResponse(streamResponse, responseTextBuilder)
 	}
 	return nil
 }
 
-func processChatCompletions(streamResp string, streamItems []string, responseTextBuilder *strings.Builder, toolCount *int) error {
-	var streamResponses []dto.ChatCompletionsStreamResponse
-	if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
-		// 一次性解析失败，逐个解析
-		common.SysLog("error unmarshalling stream response: " + err.Error())
-		for _, item := range streamItems {
-			var streamResponse dto.ChatCompletionsStreamResponse
-			if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
-				return err
-			}
-			if err := ProcessStreamResponse(streamResponse, responseTextBuilder, toolCount); err != nil {
-				common.SysLog("error processing stream response: " + err.Error())
-			}
-		}
-		return nil
+func processCompletionsStreamResponse(streamResponse dto.CompletionsStreamResponse, responseTextBuilder *strings.Builder) {
+	for _, choice := range streamResponse.Choices {
+		responseTextBuilder.WriteString(choice.Text)
 	}
-
-	// 批量处理所有响应
-	for _, streamResponse := range streamResponses {
-		for _, choice := range streamResponse.Choices {
-			responseTextBuilder.WriteString(choice.Delta.GetContentString())
-			responseTextBuilder.WriteString(choice.Delta.GetReasoningContent())
-			if choice.Delta.ToolCalls != nil {
-				if len(choice.Delta.ToolCalls) > *toolCount {
-					*toolCount = len(choice.Delta.ToolCalls)
-				}
-				for _, tool := range choice.Delta.ToolCalls {
-					responseTextBuilder.WriteString(tool.Function.Name)
-					responseTextBuilder.WriteString(tool.Function.Arguments)
-				}
-			}
-		}
-	}
-	return nil
-}
-
-func processCompletions(streamResp string, streamItems []string, responseTextBuilder *strings.Builder) error {
-	var streamResponses []dto.CompletionsStreamResponse
-	if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
-		// 一次性解析失败，逐个解析
-		common.SysLog("error unmarshalling stream response: " + err.Error())
-		for _, item := range streamItems {
-			var streamResponse dto.CompletionsStreamResponse
-			if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
-				continue
-			}
-			for _, choice := range streamResponse.Choices {
-				responseTextBuilder.WriteString(choice.Text)
-			}
-		}
-		return nil
-	}
-
-	// 批量处理所有响应
-	for _, streamResponse := range streamResponses {
-		for _, choice := range streamResponse.Choices {
-			responseTextBuilder.WriteString(choice.Text)
-		}
-	}
-	return nil
 }
 
 func handleLastResponse(lastStreamData string, responseId *string, createAt *int64,

relay/channel/openai/image_edit_test.go

@@ -0,0 +1,98 @@
+package openai
+
+import (
+	"bytes"
+	"io"
+	"mime/multipart"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/dto"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	relayconstant "github.com/QuantumNous/new-api/relay/constant"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+// TestConvertImageEditRequestMultipart verifies that ConvertImageRequest
+// re-serializes multipart image edit requests with all fields (including
+// stream) and the file intact, both when the form was already parsed and when
+// it must be re-parsed from the reusable body.
+func TestConvertImageEditRequestMultipart(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	newMultipartContext := func(t *testing.T, prompt string) *gin.Context {
+		var body bytes.Buffer
+		writer := multipart.NewWriter(&body)
+		require.NoError(t, writer.WriteField("model", "gpt-image-1"))
+		require.NoError(t, writer.WriteField("prompt", prompt))
+		require.NoError(t, writer.WriteField("stream", "true"))
+		require.NoError(t, writer.WriteField("partial_images", "3"))
+		part, err := writer.CreateFormFile("image", "input.png")
+		require.NoError(t, err)
+		_, err = part.Write([]byte("fake image"))
+		require.NoError(t, err)
+		require.NoError(t, writer.Close())
+
+		c, _ := gin.CreateTestContext(httptest.NewRecorder())
+		c.Request = httptest.NewRequest(http.MethodPost, "/v1/images/edits", &body)
+		c.Request.Header.Set("Content-Type", writer.FormDataContentType())
+		return c
+	}
+
+	convertAndReplay := func(t *testing.T, c *gin.Context, prompt string) {
+		info := &relaycommon.RelayInfo{
+			RelayMode: relayconstant.RelayModeImagesEdits,
+		}
+		request := dto.ImageRequest{
+			Model:  "gpt-image-1",
+			Prompt: prompt,
+			Stream: common.GetPointer(true),
+		}
+
+		converted, err := (&Adaptor{}).ConvertImageRequest(c, info, request)
+		require.NoError(t, err)
+		convertedBody, ok := converted.(*bytes.Buffer)
+		require.True(t, ok)
+
+		replayedRequest := httptest.NewRequest(http.MethodPost, "/v1/images/edits", bytes.NewReader(convertedBody.Bytes()))
+		replayedRequest.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
+		require.NoError(t, replayedRequest.ParseMultipartForm(32<<20))
+
+		require.Equal(t, "gpt-image-1", replayedRequest.PostForm.Get("model"))
+		require.Equal(t, prompt, replayedRequest.PostForm.Get("prompt"))
+		require.Equal(t, "true", replayedRequest.PostForm.Get("stream"))
+		require.Equal(t, "3", replayedRequest.PostForm.Get("partial_images"))
+		require.Len(t, replayedRequest.MultipartForm.File["image"], 1)
+
+		file, err := replayedRequest.MultipartForm.File["image"][0].Open()
+		require.NoError(t, err)
+		defer file.Close()
+		fileBytes, err := io.ReadAll(file)
+		require.NoError(t, err)
+		require.Equal(t, []byte("fake image"), fileBytes)
+	}
+
+	t.Run("with pre-parsed form", func(t *testing.T) {
+		prompt := "edit this image"
+		c := newMultipartContext(t, prompt)
+		require.NoError(t, c.Request.ParseMultipartForm(32<<20))
+
+		convertAndReplay(t, c, prompt)
+	})
+
+	t.Run("re-parses reusable body when form is missing", func(t *testing.T) {
+		prompt := "edit without pre-parsed form"
+		c := newMultipartContext(t, prompt)
+
+		storage, err := common.GetBodyStorage(c)
+		require.NoError(t, err)
+		c.Request.Body = io.NopCloser(storage)
+		c.Request.MultipartForm = nil
+		c.Request.PostForm = nil
+
+		convertAndReplay(t, c, prompt)
+	})
+}

relay/channel/openai/image_stream_test.go

@@ -0,0 +1,173 @@
+package openai
+
+import (
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+
+	"github.com/QuantumNous/new-api/constant"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/require"
+)
+
+func newImageTestContext(t *testing.T, body, contentType string, isStream bool) (*gin.Context, *httptest.ResponseRecorder, *http.Response, *relaycommon.RelayInfo) {
+	t.Helper()
+
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+	c.Request = httptest.NewRequest(http.MethodPost, "/v1/images/generations", nil)
+
+	resp := &http.Response{
+		StatusCode: http.StatusOK,
+		Body:       io.NopCloser(strings.NewReader(body)),
+		Header:     http.Header{"Content-Type": []string{contentType}},
+	}
+	info := &relaycommon.RelayInfo{
+		ChannelMeta: &relaycommon.ChannelMeta{},
+		IsStream:    isStream,
+	}
+	return c, recorder, resp, info
+}
+
+// TestOpenaiImageStreamHandlerForwardsSSEAndUsage covers the core SSE path:
+// chunks are forwarded with rebuilt event lines, usage is extracted and
+// normalized (input_tokens -> prompt_tokens with details), and [DONE] is
+// re-emitted to the client.
+func TestOpenaiImageStreamHandlerForwardsSSEAndUsage(t *testing.T) {
+	oldMode := gin.Mode()
+	gin.SetMode(gin.TestMode)
+	t.Cleanup(func() { gin.SetMode(oldMode) })
+
+	oldTimeout := constant.StreamingTimeout
+	constant.StreamingTimeout = 30
+	t.Cleanup(func() { constant.StreamingTimeout = oldTimeout })
+
+	body := strings.Join([]string{
+		`event: image_generation.partial_image`,
+		`data: {"type":"image_generation.partial_image","b64_json":"partial"}`,
+		``,
+		`data: {"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`,
+		``,
+		`data: [DONE]`,
+		``,
+	}, "\n")
+
+	c, recorder, resp, info := newImageTestContext(t, body, "text/event-stream", true)
+
+	usage, err := OpenaiImageStreamHandler(c, info, resp)
+	require.Nil(t, err)
+	require.Equal(t, 3, usage.PromptTokens)
+	require.Equal(t, 4, usage.CompletionTokens)
+	require.Equal(t, 7, usage.TotalTokens)
+	require.Equal(t, 2, usage.PromptTokensDetails.ImageTokens)
+	require.Equal(t, 1, usage.PromptTokensDetails.TextTokens)
+	require.Contains(t, recorder.Body.String(), `event: image_generation.partial_image`)
+	require.Contains(t, recorder.Body.String(), `data: {"type":"image_generation.partial_image","b64_json":"partial"}`)
+	require.Contains(t, recorder.Body.String(), `data: {"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`)
+	require.Contains(t, recorder.Body.String(), `data: [DONE]`)
+	require.Equal(t, "text/event-stream", recorder.Header().Get("Content-Type"))
+}
+
+// TestOpenaiImageStreamHandlerWrapsJSONResponse covers the non-SSE fallback:
+// a JSON upstream response is wrapped into pseudo-SSE completed events.
+func TestOpenaiImageStreamHandlerWrapsJSONResponse(t *testing.T) {
+	oldMode := gin.Mode()
+	gin.SetMode(gin.TestMode)
+	t.Cleanup(func() { gin.SetMode(oldMode) })
+
+	body := `{"created":1710000000,"data":[{"b64_json":"final","revised_prompt":"draw a cat"}],"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`
+
+	c, recorder, resp, info := newImageTestContext(t, body, "application/json", true)
+
+	usage, err := OpenaiImageStreamHandler(c, info, resp)
+	require.Nil(t, err)
+	require.Equal(t, 3, usage.PromptTokens)
+	require.Equal(t, 4, usage.CompletionTokens)
+	require.Equal(t, 7, usage.TotalTokens)
+	require.Equal(t, 2, usage.PromptTokensDetails.ImageTokens)
+	require.Equal(t, 1, usage.PromptTokensDetails.TextTokens)
+	require.Equal(t, "text/event-stream", recorder.Header().Get("Content-Type"))
+	require.Empty(t, recorder.Header().Get("Content-Length"))
+	require.Contains(t, recorder.Body.String(), `event: image_generation.completed`)
+	require.Contains(t, recorder.Body.String(), `"type":"image_generation.completed"`)
+	require.Contains(t, recorder.Body.String(), `"b64_json":"final"`)
+	require.Contains(t, recorder.Body.String(), `"revised_prompt":"draw a cat"`)
+	require.Contains(t, recorder.Body.String(), `data: [DONE]`)
+}
+
+// TestOpenaiImageHandlersReturnJSONError covers JSON error responses for both
+// entry points: the non-streaming handler and the stream handler's non-SSE
+// fallback. Neither must leak the error body to the client.
+func TestOpenaiImageHandlersReturnJSONError(t *testing.T) {
+	oldMode := gin.Mode()
+	gin.SetMode(gin.TestMode)
+	t.Cleanup(func() { gin.SetMode(oldMode) })
+
+	body := `{"error":{"message":"content moderation failed","type":"upstream_error","code":"content_moderation_failed","status":502}}`
+
+	t.Run("non-streaming handler", func(t *testing.T) {
+		c, recorder, resp, info := newImageTestContext(t, body, "application/json", false)
+
+		usage, err := OpenaiImageHandler(c, info, resp)
+		require.Nil(t, usage)
+		require.NotNil(t, err)
+		require.Equal(t, http.StatusOK, err.StatusCode)
+		oaiError := err.ToOpenAIError()
+		require.Equal(t, "content moderation failed", oaiError.Message)
+		require.Equal(t, "upstream_error", oaiError.Type)
+		require.Equal(t, "content_moderation_failed", oaiError.Code)
+		require.Empty(t, recorder.Body.String())
+	})
+
+	t.Run("stream handler JSON fallback", func(t *testing.T) {
+		c, recorder, resp, info := newImageTestContext(t, body, "application/json", true)
+
+		usage, err := OpenaiImageStreamHandler(c, info, resp)
+		require.Nil(t, usage)
+		require.NotNil(t, err)
+		require.Equal(t, http.StatusOK, err.StatusCode)
+		require.Equal(t, "content moderation failed", err.ToOpenAIError().Message)
+		require.Empty(t, recorder.Body.String())
+	})
+}
+
+// TestOpenaiImageStreamHandlerRecordsUpstreamErrorEvent verifies that an error
+// event inside the SSE stream is recorded as a soft error while the payload is
+// still forwarded to the client.
+func TestOpenaiImageStreamHandlerRecordsUpstreamErrorEvent(t *testing.T) {
+	oldMode := gin.Mode()
+	gin.SetMode(gin.TestMode)
+	t.Cleanup(func() { gin.SetMode(oldMode) })
+
+	oldTimeout := constant.StreamingTimeout
+	constant.StreamingTimeout = 30
+	t.Cleanup(func() { constant.StreamingTimeout = oldTimeout })
+
+	body := strings.Join([]string{
+		`event: image_generation.partial_image`,
+		`data: {"type":"image_generation.partial_image","b64_json":"partial"}`,
+		``,
+		`event: error`,
+		`data: {"type":"upstream_error","error":{"message":"stream error: stream ID 77; INTERNAL_ERROR; received from peer"}}`,
+		``,
+	}, "\n")
+
+	c, recorder, resp, info := newImageTestContext(t, body, "text/event-stream", true)
+
+	usage, err := OpenaiImageStreamHandler(c, info, resp)
+	require.Nil(t, err)
+	require.NotNil(t, usage)
+	require.NotNil(t, info.StreamStatus)
+	require.Equal(t, relaycommon.StreamEndReasonEOF, info.StreamStatus.EndReason)
+	require.True(t, info.StreamStatus.HasErrors())
+	require.Equal(t, 1, info.StreamStatus.TotalErrorCount())
+	require.Contains(t, info.StreamStatus.Errors[0].Message, "INTERNAL_ERROR")
+	// The scanner strips the upstream "event: error" line; the event name is
+	// rebuilt from the JSON "type" field (upstream_error). The error message
+	// is still forwarded in the data: payload (stream ID 77).
+	require.Contains(t, recorder.Body.String(), `event: upstream_error`)
+	require.Contains(t, recorder.Body.String(), `stream ID 77`)
+}

relay/channel/openai/relay-openai.go

@@ -14,12 +14,9 @@ import (
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
-
 	"github.com/QuantumNous/new-api/types"
 
-	"github.com/bytedance/gopkg/util/gopool"
 	"github.com/gin-gonic/gin"
-	"github.com/gorilla/websocket"
 )
 
 func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, forceFormat bool, thinkToContent bool) error {
@@ -119,7 +116,6 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 	var responseTextBuilder strings.Builder
 	var toolCount int
 	var usage = &dto.Usage{}
-	var streamItems []string // store stream items
 	var lastStreamData string
 	var secondLastStreamData string // 存储倒数第二个stream data，用于音频模型
 
@@ -140,7 +136,10 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 			}
 
 			lastStreamData = data
-			streamItems = append(streamItems, data)
+			if err := processTokenData(info.RelayMode, data, &responseTextBuilder, &toolCount); err != nil {
+				logger.LogError(c, "error processing stream token data: "+err.Error())
+				sr.Error(err)
+			}
 		}
 	})
 
@@ -155,9 +154,9 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 			containStreamUsage = true
 
 			if common.DebugEnabled {
-				logger.LogDebug(c, fmt.Sprintf("Audio model usage extracted from second last SSE: PromptTokens=%d, CompletionTokens=%d, TotalTokens=%d, InputTokens=%d, OutputTokens=%d",
+				logger.LogDebug(c, "Audio model usage extracted from second last SSE: PromptTokens=%d, CompletionTokens=%d, TotalTokens=%d, InputTokens=%d, OutputTokens=%d",
 					usage.PromptTokens, usage.CompletionTokens, usage.TotalTokens,
-					usage.InputTokens, usage.OutputTokens))
+					usage.InputTokens, usage.OutputTokens)
 			}
 		}
 	}
@@ -175,11 +174,6 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 		}
 	}
 
-	// 处理token计算
-	if err := processTokens(info.RelayMode, streamItems, &responseTextBuilder, &toolCount); err != nil {
-		logger.LogError(c, "error processing tokens: "+err.Error())
-	}
-
 	if !containStreamUsage {
 		usage = service.ResponseText2Usage(c, responseTextBuilder.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
 		usage.CompletionTokens += toolCount * 7
@@ -200,9 +194,7 @@ func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 	if err != nil {
 		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
 	}
-	if common.DebugEnabled {
-		println("upstream response body:", string(responseBody))
-	}
+	logger.LogDebug(c, "upstream response body: %s", responseBody)
 	// Unmarshal to simpleResponse
 	if info.ChannelType == constant.ChannelTypeOpenRouter && info.ChannelOtherSettings.IsOpenRouterEnterprise() {
 		// 尝试解析为 openrouter enterprise
@@ -298,421 +290,3 @@ func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 
 	return &simpleResponse.Usage, nil
 }
-
-func streamTTSResponse(c *gin.Context, resp *http.Response) {
-	c.Writer.WriteHeaderNow()
-
-	flusher, ok := c.Writer.(http.Flusher)
-	if !ok {
-		logger.LogWarn(c, "streaming not supported")
-		_, err := io.Copy(c.Writer, resp.Body)
-		if err != nil {
-			logger.LogWarn(c, err.Error())
-		}
-		return
-	}
-
-	buffer := make([]byte, 4096)
-	for {
-		n, err := resp.Body.Read(buffer)
-		//logger.LogInfo(c, fmt.Sprintf("streamTTSResponse read %d bytes", n))
-		if n > 0 {
-			if _, writeErr := c.Writer.Write(buffer[:n]); writeErr != nil {
-				logger.LogError(c, writeErr.Error())
-				break
-			}
-			flusher.Flush()
-		}
-		if err != nil {
-			if err != io.EOF {
-				logger.LogError(c, err.Error())
-			}
-			break
-		}
-	}
-}
-
-func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.NewAPIError, *dto.RealtimeUsage) {
-	if info == nil || info.ClientWs == nil || info.TargetWs == nil {
-		return types.NewError(fmt.Errorf("invalid websocket connection"), types.ErrorCodeBadResponse), nil
-	}
-
-	info.IsStream = true
-	clientConn := info.ClientWs
-	targetConn := info.TargetWs
-
-	clientClosed := make(chan struct{})
-	targetClosed := make(chan struct{})
-	sendChan := make(chan []byte, 100)
-	receiveChan := make(chan []byte, 100)
-	errChan := make(chan error, 2)
-
-	usage := &dto.RealtimeUsage{}
-	localUsage := &dto.RealtimeUsage{}
-	sumUsage := &dto.RealtimeUsage{}
-
-	gopool.Go(func() {
-		defer func() {
-			if r := recover(); r != nil {
-				errChan <- fmt.Errorf("panic in client reader: %v", r)
-			}
-		}()
-		for {
-			select {
-			case <-c.Done():
-				return
-			default:
-				_, message, err := clientConn.ReadMessage()
-				if err != nil {
-					if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
-						errChan <- fmt.Errorf("error reading from client: %v", err)
-					}
-					close(clientClosed)
-					return
-				}
-
-				realtimeEvent := &dto.RealtimeEvent{}
-				err = common.Unmarshal(message, realtimeEvent)
-				if err != nil {
-					errChan <- fmt.Errorf("error unmarshalling message: %v", err)
-					return
-				}
-
-				if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdate {
-					if realtimeEvent.Session != nil {
-						if realtimeEvent.Session.Tools != nil {
-							info.RealtimeTools = realtimeEvent.Session.Tools
-						}
-					}
-				}
-
-				textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
-				if err != nil {
-					errChan <- fmt.Errorf("error counting text token: %v", err)
-					return
-				}
-				logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
-				localUsage.TotalTokens += textToken + audioToken
-				localUsage.InputTokens += textToken + audioToken
-				localUsage.InputTokenDetails.TextTokens += textToken
-				localUsage.InputTokenDetails.AudioTokens += audioToken
-
-				err = helper.WssString(c, targetConn, string(message))
-				if err != nil {
-					errChan <- fmt.Errorf("error writing to target: %v", err)
-					return
-				}
-
-				select {
-				case sendChan <- message:
-				default:
-				}
-			}
-		}
-	})
-
-	gopool.Go(func() {
-		defer func() {
-			if r := recover(); r != nil {
-				errChan <- fmt.Errorf("panic in target reader: %v", r)
-			}
-		}()
-		for {
-			select {
-			case <-c.Done():
-				return
-			default:
-				_, message, err := targetConn.ReadMessage()
-				if err != nil {
-					if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
-						errChan <- fmt.Errorf("error reading from target: %v", err)
-					}
-					close(targetClosed)
-					return
-				}
-				info.SetFirstResponseTime()
-				realtimeEvent := &dto.RealtimeEvent{}
-				err = common.Unmarshal(message, realtimeEvent)
-				if err != nil {
-					errChan <- fmt.Errorf("error unmarshalling message: %v", err)
-					return
-				}
-
-				if realtimeEvent.Type == dto.RealtimeEventTypeResponseDone {
-					realtimeUsage := realtimeEvent.Response.Usage
-					if realtimeUsage != nil {
-						usage.TotalTokens += realtimeUsage.TotalTokens
-						usage.InputTokens += realtimeUsage.InputTokens
-						usage.OutputTokens += realtimeUsage.OutputTokens
-						usage.InputTokenDetails.AudioTokens += realtimeUsage.InputTokenDetails.AudioTokens
-						usage.InputTokenDetails.CachedTokens += realtimeUsage.InputTokenDetails.CachedTokens
-						usage.InputTokenDetails.TextTokens += realtimeUsage.InputTokenDetails.TextTokens
-						usage.OutputTokenDetails.AudioTokens += realtimeUsage.OutputTokenDetails.AudioTokens
-						usage.OutputTokenDetails.TextTokens += realtimeUsage.OutputTokenDetails.TextTokens
-						err := preConsumeUsage(c, info, usage, sumUsage)
-						if err != nil {
-							errChan <- fmt.Errorf("error consume usage: %v", err)
-							return
-						}
-						// 本次计费完成，清除
-						usage = &dto.RealtimeUsage{}
-
-						localUsage = &dto.RealtimeUsage{}
-					} else {
-						textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
-						if err != nil {
-							errChan <- fmt.Errorf("error counting text token: %v", err)
-							return
-						}
-						logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
-						localUsage.TotalTokens += textToken + audioToken
-						info.IsFirstRequest = false
-						localUsage.InputTokens += textToken + audioToken
-						localUsage.InputTokenDetails.TextTokens += textToken
-						localUsage.InputTokenDetails.AudioTokens += audioToken
-						err = preConsumeUsage(c, info, localUsage, sumUsage)
-						if err != nil {
-							errChan <- fmt.Errorf("error consume usage: %v", err)
-							return
-						}
-						// 本次计费完成，清除
-						localUsage = &dto.RealtimeUsage{}
-						// print now usage
-					}
-					logger.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
-					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
-					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
-
-				} else if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdated || realtimeEvent.Type == dto.RealtimeEventTypeSessionCreated {
-					realtimeSession := realtimeEvent.Session
-					if realtimeSession != nil {
-						// update audio format
-						info.InputAudioFormat = common.GetStringIfEmpty(realtimeSession.InputAudioFormat, info.InputAudioFormat)
-						info.OutputAudioFormat = common.GetStringIfEmpty(realtimeSession.OutputAudioFormat, info.OutputAudioFormat)
-					}
-				} else {
-					textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
-					if err != nil {
-						errChan <- fmt.Errorf("error counting text token: %v", err)
-						return
-					}
-					logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
-					localUsage.TotalTokens += textToken + audioToken
-					localUsage.OutputTokens += textToken + audioToken
-					localUsage.OutputTokenDetails.TextTokens += textToken
-					localUsage.OutputTokenDetails.AudioTokens += audioToken
-				}
-
-				err = helper.WssString(c, clientConn, string(message))
-				if err != nil {
-					errChan <- fmt.Errorf("error writing to client: %v", err)
-					return
-				}
-
-				select {
-				case receiveChan <- message:
-				default:
-				}
-			}
-		}
-	})
-
-	select {
-	case <-clientClosed:
-	case <-targetClosed:
-	case err := <-errChan:
-		//return service.OpenAIErrorWrapper(err, "realtime_error", http.StatusInternalServerError), nil
-		logger.LogError(c, "realtime error: "+err.Error())
-	case <-c.Done():
-	}
-
-	if usage.TotalTokens != 0 {
-		_ = preConsumeUsage(c, info, usage, sumUsage)
-	}
-
-	if localUsage.TotalTokens != 0 {
-		_ = preConsumeUsage(c, info, localUsage, sumUsage)
-	}
-
-	// check usage total tokens, if 0, use local usage
-
-	return nil, sumUsage
-}
-
-func preConsumeUsage(ctx *gin.Context, info *relaycommon.RelayInfo, usage *dto.RealtimeUsage, totalUsage *dto.RealtimeUsage) error {
-	if usage == nil || totalUsage == nil {
-		return fmt.Errorf("invalid usage pointer")
-	}
-
-	totalUsage.TotalTokens += usage.TotalTokens
-	totalUsage.InputTokens += usage.InputTokens
-	totalUsage.OutputTokens += usage.OutputTokens
-	totalUsage.InputTokenDetails.CachedTokens += usage.InputTokenDetails.CachedTokens
-	totalUsage.InputTokenDetails.TextTokens += usage.InputTokenDetails.TextTokens
-	totalUsage.InputTokenDetails.AudioTokens += usage.InputTokenDetails.AudioTokens
-	totalUsage.OutputTokenDetails.TextTokens += usage.OutputTokenDetails.TextTokens
-	totalUsage.OutputTokenDetails.AudioTokens += usage.OutputTokenDetails.AudioTokens
-	// clear usage
-	err := service.PreWssConsumeQuota(ctx, info, usage)
-	return err
-}
-
-func OpenaiHandlerWithUsage(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	defer service.CloseResponseBodyGracefully(resp)
-
-	responseBody, err := io.ReadAll(resp.Body)
-	if err != nil {
-		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
-	}
-
-	var usageResp dto.SimpleResponse
-	err = common.Unmarshal(responseBody, &usageResp)
-	if err != nil {
-		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
-	}
-
-	// 写入新的 response body
-	service.IOCopyBytesGracefully(c, resp, responseBody)
-
-	// Once we've written to the client, we should not return errors anymore
-	// because the upstream has already consumed resources and returned content
-	// We should still perform billing even if parsing fails
-	// format
-	if usageResp.InputTokens > 0 {
-		usageResp.PromptTokens += usageResp.InputTokens
-	}
-	if usageResp.OutputTokens > 0 {
-		usageResp.CompletionTokens += usageResp.OutputTokens
-	}
-	if usageResp.InputTokensDetails != nil {
-		usageResp.PromptTokensDetails.ImageTokens += usageResp.InputTokensDetails.ImageTokens
-		usageResp.PromptTokensDetails.TextTokens += usageResp.InputTokensDetails.TextTokens
-	}
-	applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
-	return &usageResp.Usage, nil
-}
-
-func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, responseBody []byte) {
-	if info == nil || usage == nil {
-		return
-	}
-
-	switch info.ChannelType {
-	case constant.ChannelTypeDeepSeek:
-		if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
-			usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
-		}
-	case constant.ChannelTypeZhipu_v4:
-		// 智普的cached_tokens在标准位置: usage.prompt_tokens_details.cached_tokens
-		if usage.PromptTokensDetails.CachedTokens == 0 {
-			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
-				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
-			} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
-				usage.PromptTokensDetails.CachedTokens = cachedTokens
-			} else if usage.PromptCacheHitTokens > 0 {
-				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
-			}
-		}
-	case constant.ChannelTypeMoonshot:
-		// Moonshot的cached_tokens在非标准位置: choices[].usage.cached_tokens
-		if usage.PromptTokensDetails.CachedTokens == 0 {
-			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
-				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
-			} else if cachedTokens, ok := extractMoonshotCachedTokensFromBody(responseBody); ok {
-				usage.PromptTokensDetails.CachedTokens = cachedTokens
-			} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
-				usage.PromptTokensDetails.CachedTokens = cachedTokens
-			} else if usage.PromptCacheHitTokens > 0 {
-				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
-			}
-		}
-	case constant.ChannelTypeOpenAI:
-		if usage.PromptTokensDetails.CachedTokens == 0 {
-			if cachedTokens, ok := extractLlamaCachedTokensFromBody(responseBody); ok {
-				usage.PromptTokensDetails.CachedTokens = cachedTokens
-			}
-		}
-	}
-}
-
-func extractCachedTokensFromBody(body []byte) (int, bool) {
-	if len(body) == 0 {
-		return 0, false
-	}
-
-	var payload struct {
-		Usage struct {
-			PromptTokensDetails struct {
-				CachedTokens *int `json:"cached_tokens"`
-			} `json:"prompt_tokens_details"`
-			CachedTokens         *int `json:"cached_tokens"`
-			PromptCacheHitTokens *int `json:"prompt_cache_hit_tokens"`
-		} `json:"usage"`
-	}
-
-	if err := common.Unmarshal(body, &payload); err != nil {
-		return 0, false
-	}
-
-	if payload.Usage.PromptTokensDetails.CachedTokens != nil {
-		return *payload.Usage.PromptTokensDetails.CachedTokens, true
-	}
-	if payload.Usage.CachedTokens != nil {
-		return *payload.Usage.CachedTokens, true
-	}
-	if payload.Usage.PromptCacheHitTokens != nil {
-		return *payload.Usage.PromptCacheHitTokens, true
-	}
-	return 0, false
-}
-
-// extractMoonshotCachedTokensFromBody 从Moonshot的非标准位置提取cached_tokens
-// Moonshot的流式响应格式: {"choices":[{"usage":{"cached_tokens":111}}]}
-func extractMoonshotCachedTokensFromBody(body []byte) (int, bool) {
-	if len(body) == 0 {
-		return 0, false
-	}
-
-	var payload struct {
-		Choices []struct {
-			Usage struct {
-				CachedTokens *int `json:"cached_tokens"`
-			} `json:"usage"`
-		} `json:"choices"`
-	}
-
-	if err := common.Unmarshal(body, &payload); err != nil {
-		return 0, false
-	}
-
-	// 遍历choices查找cached_tokens
-	for _, choice := range payload.Choices {
-		if choice.Usage.CachedTokens != nil && *choice.Usage.CachedTokens > 0 {
-			return *choice.Usage.CachedTokens, true
-		}
-	}
-
-	return 0, false
-}
-
-// extractLlamaCachedTokensFromBody 从llama.cpp的非标准位置提取cache_n
-func extractLlamaCachedTokensFromBody(body []byte) (int, bool) {
-	if len(body) == 0 {
-		return 0, false
-	}
-
-	var payload struct {
-		Timings struct {
-			CachedTokens *int `json:"cache_n"`
-		} `json:"timings"`
-	}
-
-	if err := common.Unmarshal(body, &payload); err != nil {
-		return 0, false
-	}
-
-	if payload.Timings.CachedTokens == nil {
-		return 0, false
-	}
-	return *payload.Timings.CachedTokens, true
-}

relay/channel/openai/relay_image.go

@@ -0,0 +1,287 @@
+package openai
+
+import (
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"strings"
+	"time"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/QuantumNous/new-api/relay/helper"
+	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/types"
+
+	"github.com/gin-gonic/gin"
+)
+
+// OpenaiImageHandler handles non-streaming OpenAI image responses
+// (generations/edits), returning the parsed usage for billing.
+func OpenaiImageHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
+	defer service.CloseResponseBodyGracefully(resp)
+
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
+	}
+
+	var usageResp dto.SimpleResponse
+	err = common.Unmarshal(responseBody, &usageResp)
+	if err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
+	}
+
+	if oaiError := usageResp.GetOpenAIError(); oaiError != nil && oaiError.Type != "" {
+		return nil, types.WithOpenAIError(*oaiError, resp.StatusCode)
+	}
+
+	// 写入新的 response body
+	service.IOCopyBytesGracefully(c, resp, responseBody)
+
+	normalizeOpenAIUsage(&usageResp.Usage)
+	applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
+	return &usageResp.Usage, nil
+}
+
+// normalizeOpenAIUsage maps the OpenAI Images usage shape (input_tokens /
+// output_tokens / input_tokens_details) onto the canonical prompt/completion
+// fields. It is used only on the OpenAI image relay paths (generations/edits,
+// streaming and non-streaming): the image API never returns prompt_tokens /
+// completion_tokens, so the overwrite (=) semantics here are equivalent to the
+// previous additive (+=) behavior while avoiding any future double-counting if
+// both field sets are ever populated. Do not reuse this on chat/embedding paths
+// without revisiting the overwrite semantics.
+func normalizeOpenAIUsage(usage *dto.Usage) {
+	if usage == nil {
+		return
+	}
+	if usage.InputTokens != 0 {
+		usage.PromptTokens = usage.InputTokens
+	}
+	if usage.OutputTokens != 0 {
+		usage.CompletionTokens = usage.OutputTokens
+	}
+	if usage.InputTokensDetails != nil {
+		usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
+		usage.PromptTokensDetails.CachedCreationTokens = usage.InputTokensDetails.CachedCreationTokens
+		usage.PromptTokensDetails.ImageTokens = usage.InputTokensDetails.ImageTokens
+		usage.PromptTokensDetails.TextTokens = usage.InputTokensDetails.TextTokens
+		usage.PromptTokensDetails.AudioTokens = usage.InputTokensDetails.AudioTokens
+	}
+	if usage.TotalTokens == 0 {
+		usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
+	}
+}
+
+func OpenaiImageStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
+	if resp == nil || resp.Body == nil {
+		logger.LogError(c, "invalid image stream response")
+		return nil, types.NewOpenAIError(fmt.Errorf("invalid response"), types.ErrorCodeBadResponse, http.StatusInternalServerError)
+	}
+
+	contentType := strings.ToLower(resp.Header.Get("Content-Type"))
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
+		return OpenaiImageHandler(c, info, resp)
+	}
+	if !strings.Contains(contentType, "text/event-stream") {
+		return OpenaiImageJSONAsStreamHandler(c, info, resp)
+	}
+	// Reuse the shared streaming engine (helper.StreamScannerHandler) so the
+	// image streaming path gets the same ping keepalive, streaming-timeout
+	// watchdog, client-disconnect detection, panic recovery and goroutine
+	// cleanup as every other relay stream. The scanner delivers only the
+	// "data:" payload, so the SSE "event:" line is rebuilt from the JSON "type"
+	// field (real OpenAI image events keep event == type).
+	usage := &dto.Usage{}
+	var lastStreamData []byte
+
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
+		raw := common.StringToByteSlice(data)
+		lastStreamData = raw
+		if isOpenAIImageStreamErrorEvent(raw) {
+			// Record the error as a soft error; the scanner drives the final
+			// EndReason. HasErrors() flags the failure for logging/handling.
+			sr.Error(fmt.Errorf("%s", extractOpenAIImageStreamErrorMessage(raw)))
+		}
+		var usageResp dto.SimpleResponse
+		if err := common.Unmarshal(raw, &usageResp); err == nil {
+			normalizeOpenAIUsage(&usageResp.Usage)
+			if service.ValidUsage(&usageResp.Usage) {
+				usage = &usageResp.Usage
+			}
+		}
+		writeOpenaiImageStreamChunk(c, raw)
+	})
+
+	// StreamScannerHandler consumes the upstream [DONE]; re-emit it so the
+	// client still receives a terminal data: [DONE].
+	if info != nil && info.StreamStatus != nil && info.StreamStatus.EndReason == relaycommon.StreamEndReasonDone {
+		helper.Done(c)
+	}
+
+	applyUsagePostProcessing(info, usage, lastStreamData)
+	return usage, nil
+}
+
+// writeOpenaiImageStreamChunk rebuilds the SSE frame for an image stream chunk:
+// it emits an "event:" line derived from the JSON "type" field (when present)
+// followed by the verbatim "data:" payload, mirroring helper.ResponseChunkData.
+func writeOpenaiImageStreamChunk(c *gin.Context, data []byte) {
+	var payload struct {
+		Type string `json:"type"`
+	}
+	_ = common.Unmarshal(data, &payload)
+	if eventName := strings.TrimSpace(payload.Type); eventName != "" {
+		c.Render(-1, common.CustomEvent{Data: fmt.Sprintf("event: %s\n", eventName)})
+	}
+	c.Render(-1, common.CustomEvent{Data: "data: " + string(data)})
+	_ = helper.FlushWriter(c)
+}
+
+// isOpenAIImageStreamErrorEvent detects upstream error chunks by JSON content
+// only ("type" of error/upstream_error, or a non-empty "error" field). The SSE
+// "event:" line is not available here: StreamScannerHandler delivers only the
+// "data:" payload. A payload carrying just a "message" key is deliberately NOT
+// treated as an error to avoid false positives.
+func isOpenAIImageStreamErrorEvent(data []byte) bool {
+	if !json.Valid(data) {
+		return false
+	}
+	var payload struct {
+		Type  string          `json:"type"`
+		Error json.RawMessage `json:"error"`
+	}
+	if err := common.Unmarshal(data, &payload); err != nil {
+		return false
+	}
+	payloadType := strings.ToLower(strings.TrimSpace(payload.Type))
+	return payloadType == "error" || payloadType == "upstream_error" || len(payload.Error) > 0
+}
+
+func extractOpenAIImageStreamErrorMessage(data []byte) string {
+	if len(data) == 0 || !json.Valid(data) {
+		return "upstream image stream returned error event"
+	}
+	var payload struct {
+		Message string          `json:"message"`
+		Error   json.RawMessage `json:"error"`
+	}
+	if err := common.Unmarshal(data, &payload); err != nil {
+		return "upstream image stream returned error event"
+	}
+	if msg := strings.TrimSpace(payload.Message); msg != "" {
+		return msg
+	}
+	if len(payload.Error) > 0 {
+		var nested struct {
+			Message string `json:"message"`
+		}
+		if err := common.Unmarshal(payload.Error, &nested); err == nil {
+			if msg := strings.TrimSpace(nested.Message); msg != "" {
+				return msg
+			}
+		}
+		if msg := strings.TrimSpace(common.JsonRawMessageToString(payload.Error)); msg != "" {
+			return msg
+		}
+	}
+	return "upstream image stream returned error event"
+}
+
+func OpenaiImageJSONAsStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
+	defer service.CloseResponseBodyGracefully(resp)
+
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
+	}
+
+	var imageResp dto.ImageResponse
+	if err := common.Unmarshal(responseBody, &imageResp); err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
+	}
+
+	var usageResp dto.SimpleResponse
+	_ = common.Unmarshal(responseBody, &usageResp)
+	if oaiError := usageResp.GetOpenAIError(); oaiError != nil && oaiError.Type != "" {
+		return nil, types.WithOpenAIError(*oaiError, resp.StatusCode)
+	}
+	normalizeOpenAIUsage(&usageResp.Usage)
+	applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
+
+	helper.SetEventStreamHeaders(c)
+	c.Status(http.StatusOK)
+
+	created := imageResp.Created
+	if created == 0 {
+		created = time.Now().Unix()
+	}
+	if info != nil {
+		info.SetFirstResponseTime()
+	}
+	for _, image := range imageResp.Data {
+		payload := map[string]any{
+			"type":       "image_generation.completed",
+			"created_at": created,
+		}
+		if image.Url != "" {
+			payload["url"] = image.Url
+		}
+		if image.B64Json != "" {
+			payload["b64_json"] = image.B64Json
+		}
+		if image.RevisedPrompt != "" {
+			payload["revised_prompt"] = image.RevisedPrompt
+		}
+		if service.ValidUsage(&usageResp.Usage) {
+			payload["usage"] = usageResp.Usage
+		}
+		if err := writeOpenaiImageStreamPayload(c, "image_generation.completed", payload); err != nil {
+			if info != nil && info.StreamStatus != nil {
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, err)
+			}
+			return &usageResp.Usage, nil
+		}
+	}
+	if err := writeOpenaiImageStreamDone(c); err != nil {
+		if info != nil && info.StreamStatus != nil {
+			info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, err)
+		}
+		return &usageResp.Usage, nil
+	}
+	if info != nil {
+		info.ReceivedResponseCount += len(imageResp.Data)
+		if info.StreamStatus == nil {
+			info.StreamStatus = relaycommon.NewStreamStatus()
+		}
+		info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonDone, nil)
+	}
+	return &usageResp.Usage, nil
+}
+
+func writeOpenaiImageStreamPayload(c *gin.Context, eventName string, payload any) error {
+	data, err := common.Marshal(payload)
+	if err != nil {
+		return err
+	}
+	if eventName != "" {
+		if _, err := fmt.Fprintf(c.Writer, "event: %s\n", eventName); err != nil {
+			return err
+		}
+	}
+	if _, err := fmt.Fprintf(c.Writer, "data: %s\n\n", data); err != nil {
+		return err
+	}
+	return helper.FlushWriter(c)
+}
+
+func writeOpenaiImageStreamDone(c *gin.Context) error {
+	if _, err := fmt.Fprint(c.Writer, "data: [DONE]\n\n"); err != nil {
+		return err
+	}
+	return helper.FlushWriter(c)
+}

relay/channel/openai/relay_realtime.go

@@ -0,0 +1,242 @@
+package openai
+
+import (
+	"fmt"
+
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	"github.com/QuantumNous/new-api/relay/helper"
+	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/types"
+
+	"github.com/bytedance/gopkg/util/gopool"
+	"github.com/gin-gonic/gin"
+	"github.com/gorilla/websocket"
+)
+
+func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.NewAPIError, *dto.RealtimeUsage) {
+	if info == nil || info.ClientWs == nil || info.TargetWs == nil {
+		return types.NewError(fmt.Errorf("invalid websocket connection"), types.ErrorCodeBadResponse), nil
+	}
+
+	info.IsStream = true
+	clientConn := info.ClientWs
+	targetConn := info.TargetWs
+
+	clientClosed := make(chan struct{})
+	targetClosed := make(chan struct{})
+	sendChan := make(chan []byte, 100)
+	receiveChan := make(chan []byte, 100)
+	errChan := make(chan error, 2)
+
+	usage := &dto.RealtimeUsage{}
+	localUsage := &dto.RealtimeUsage{}
+	sumUsage := &dto.RealtimeUsage{}
+
+	gopool.Go(func() {
+		defer func() {
+			if r := recover(); r != nil {
+				errChan <- fmt.Errorf("panic in client reader: %v", r)
+			}
+		}()
+		for {
+			select {
+			case <-c.Done():
+				return
+			default:
+				_, message, err := clientConn.ReadMessage()
+				if err != nil {
+					if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
+						errChan <- fmt.Errorf("error reading from client: %v", err)
+					}
+					close(clientClosed)
+					return
+				}
+
+				realtimeEvent := &dto.RealtimeEvent{}
+				err = common.Unmarshal(message, realtimeEvent)
+				if err != nil {
+					errChan <- fmt.Errorf("error unmarshalling message: %v", err)
+					return
+				}
+
+				if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdate {
+					if realtimeEvent.Session != nil {
+						if realtimeEvent.Session.Tools != nil {
+							info.RealtimeTools = realtimeEvent.Session.Tools
+						}
+					}
+				}
+
+				textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
+				if err != nil {
+					errChan <- fmt.Errorf("error counting text token: %v", err)
+					return
+				}
+				logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+				localUsage.TotalTokens += textToken + audioToken
+				localUsage.InputTokens += textToken + audioToken
+				localUsage.InputTokenDetails.TextTokens += textToken
+				localUsage.InputTokenDetails.AudioTokens += audioToken
+
+				err = helper.WssString(c, targetConn, string(message))
+				if err != nil {
+					errChan <- fmt.Errorf("error writing to target: %v", err)
+					return
+				}
+
+				select {
+				case sendChan <- message:
+				default:
+				}
+			}
+		}
+	})
+
+	gopool.Go(func() {
+		defer func() {
+			if r := recover(); r != nil {
+				errChan <- fmt.Errorf("panic in target reader: %v", r)
+			}
+		}()
+		for {
+			select {
+			case <-c.Done():
+				return
+			default:
+				_, message, err := targetConn.ReadMessage()
+				if err != nil {
+					if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
+						errChan <- fmt.Errorf("error reading from target: %v", err)
+					}
+					close(targetClosed)
+					return
+				}
+				info.SetFirstResponseTime()
+				realtimeEvent := &dto.RealtimeEvent{}
+				err = common.Unmarshal(message, realtimeEvent)
+				if err != nil {
+					errChan <- fmt.Errorf("error unmarshalling message: %v", err)
+					return
+				}
+
+				if realtimeEvent.Type == dto.RealtimeEventTypeResponseDone {
+					realtimeUsage := realtimeEvent.Response.Usage
+					if realtimeUsage != nil {
+						usage.TotalTokens += realtimeUsage.TotalTokens
+						usage.InputTokens += realtimeUsage.InputTokens
+						usage.OutputTokens += realtimeUsage.OutputTokens
+						usage.InputTokenDetails.AudioTokens += realtimeUsage.InputTokenDetails.AudioTokens
+						usage.InputTokenDetails.CachedTokens += realtimeUsage.InputTokenDetails.CachedTokens
+						usage.InputTokenDetails.TextTokens += realtimeUsage.InputTokenDetails.TextTokens
+						usage.OutputTokenDetails.AudioTokens += realtimeUsage.OutputTokenDetails.AudioTokens
+						usage.OutputTokenDetails.TextTokens += realtimeUsage.OutputTokenDetails.TextTokens
+						err := preConsumeUsage(c, info, usage, sumUsage)
+						if err != nil {
+							errChan <- fmt.Errorf("error consume usage: %v", err)
+							return
+						}
+						// 本次计费完成，清除
+						usage = &dto.RealtimeUsage{}
+
+						localUsage = &dto.RealtimeUsage{}
+					} else {
+						textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
+						if err != nil {
+							errChan <- fmt.Errorf("error counting text token: %v", err)
+							return
+						}
+						logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+						localUsage.TotalTokens += textToken + audioToken
+						info.IsFirstRequest = false
+						localUsage.InputTokens += textToken + audioToken
+						localUsage.InputTokenDetails.TextTokens += textToken
+						localUsage.InputTokenDetails.AudioTokens += audioToken
+						err = preConsumeUsage(c, info, localUsage, sumUsage)
+						if err != nil {
+							errChan <- fmt.Errorf("error consume usage: %v", err)
+							return
+						}
+						// 本次计费完成，清除
+						localUsage = &dto.RealtimeUsage{}
+						// print now usage
+					}
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
+
+				} else if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdated || realtimeEvent.Type == dto.RealtimeEventTypeSessionCreated {
+					realtimeSession := realtimeEvent.Session
+					if realtimeSession != nil {
+						// update audio format
+						info.InputAudioFormat = common.GetStringIfEmpty(realtimeSession.InputAudioFormat, info.InputAudioFormat)
+						info.OutputAudioFormat = common.GetStringIfEmpty(realtimeSession.OutputAudioFormat, info.OutputAudioFormat)
+					}
+				} else {
+					textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
+					if err != nil {
+						errChan <- fmt.Errorf("error counting text token: %v", err)
+						return
+					}
+					logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+					localUsage.TotalTokens += textToken + audioToken
+					localUsage.OutputTokens += textToken + audioToken
+					localUsage.OutputTokenDetails.TextTokens += textToken
+					localUsage.OutputTokenDetails.AudioTokens += audioToken
+				}
+
+				err = helper.WssString(c, clientConn, string(message))
+				if err != nil {
+					errChan <- fmt.Errorf("error writing to client: %v", err)
+					return
+				}
+
+				select {
+				case receiveChan <- message:
+				default:
+				}
+			}
+		}
+	})
+
+	select {
+	case <-clientClosed:
+	case <-targetClosed:
+	case err := <-errChan:
+		//return service.OpenAIErrorWrapper(err, "realtime_error", http.StatusInternalServerError), nil
+		logger.LogError(c, "realtime error: "+err.Error())
+	case <-c.Done():
+	}
+
+	if usage.TotalTokens != 0 {
+		_ = preConsumeUsage(c, info, usage, sumUsage)
+	}
+
+	if localUsage.TotalTokens != 0 {
+		_ = preConsumeUsage(c, info, localUsage, sumUsage)
+	}
+
+	// check usage total tokens, if 0, use local usage
+
+	return nil, sumUsage
+}
+
+func preConsumeUsage(ctx *gin.Context, info *relaycommon.RelayInfo, usage *dto.RealtimeUsage, totalUsage *dto.RealtimeUsage) error {
+	if usage == nil || totalUsage == nil {
+		return fmt.Errorf("invalid usage pointer")
+	}
+
+	totalUsage.TotalTokens += usage.TotalTokens
+	totalUsage.InputTokens += usage.InputTokens
+	totalUsage.OutputTokens += usage.OutputTokens
+	totalUsage.InputTokenDetails.CachedTokens += usage.InputTokenDetails.CachedTokens
+	totalUsage.InputTokenDetails.TextTokens += usage.InputTokenDetails.TextTokens
+	totalUsage.InputTokenDetails.AudioTokens += usage.InputTokenDetails.AudioTokens
+	totalUsage.OutputTokenDetails.TextTokens += usage.OutputTokenDetails.TextTokens
+	totalUsage.OutputTokenDetails.AudioTokens += usage.OutputTokenDetails.AudioTokens
+	// clear usage
+	err := service.PreWssConsumeQuota(ctx, info, usage)
+	return err
+}

relay/channel/openai/usage.go

@@ -0,0 +1,133 @@
+package openai
+
+import (
+	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/dto"
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+)
+
+func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, responseBody []byte) {
+	if info == nil || usage == nil {
+		return
+	}
+
+	switch info.ChannelType {
+	case constant.ChannelTypeDeepSeek:
+		if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
+			usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
+		}
+	case constant.ChannelTypeZhipu_v4:
+		// 智普的cached_tokens在标准位置: usage.prompt_tokens_details.cached_tokens
+		if usage.PromptTokensDetails.CachedTokens == 0 {
+			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
+				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
+			} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
+				usage.PromptTokensDetails.CachedTokens = cachedTokens
+			} else if usage.PromptCacheHitTokens > 0 {
+				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
+			}
+		}
+	case constant.ChannelTypeMoonshot:
+		// Moonshot的cached_tokens在非标准位置: choices[].usage.cached_tokens
+		if usage.PromptTokensDetails.CachedTokens == 0 {
+			if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
+				usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
+			} else if cachedTokens, ok := extractMoonshotCachedTokensFromBody(responseBody); ok {
+				usage.PromptTokensDetails.CachedTokens = cachedTokens
+			} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
+				usage.PromptTokensDetails.CachedTokens = cachedTokens
+			} else if usage.PromptCacheHitTokens > 0 {
+				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
+			}
+		}
+	case constant.ChannelTypeOpenAI:
+		if usage.PromptTokensDetails.CachedTokens == 0 {
+			if cachedTokens, ok := extractLlamaCachedTokensFromBody(responseBody); ok {
+				usage.PromptTokensDetails.CachedTokens = cachedTokens
+			}
+		}
+	}
+}
+
+func extractCachedTokensFromBody(body []byte) (int, bool) {
+	if len(body) == 0 {
+		return 0, false
+	}
+
+	var payload struct {
+		Usage struct {
+			PromptTokensDetails struct {
+				CachedTokens *int `json:"cached_tokens"`
+			} `json:"prompt_tokens_details"`
+			CachedTokens         *int `json:"cached_tokens"`
+			PromptCacheHitTokens *int `json:"prompt_cache_hit_tokens"`
+		} `json:"usage"`
+	}
+
+	if err := common.Unmarshal(body, &payload); err != nil {
+		return 0, false
+	}
+
+	if payload.Usage.PromptTokensDetails.CachedTokens != nil {
+		return *payload.Usage.PromptTokensDetails.CachedTokens, true
+	}
+	if payload.Usage.CachedTokens != nil {
+		return *payload.Usage.CachedTokens, true
+	}
+	if payload.Usage.PromptCacheHitTokens != nil {
+		return *payload.Usage.PromptCacheHitTokens, true
+	}
+	return 0, false
+}
+
+// extractMoonshotCachedTokensFromBody 从Moonshot的非标准位置提取cached_tokens
+// Moonshot的流式响应格式: {"choices":[{"usage":{"cached_tokens":111}}]}
+func extractMoonshotCachedTokensFromBody(body []byte) (int, bool) {
+	if len(body) == 0 {
+		return 0, false
+	}
+
+	var payload struct {
+		Choices []struct {
+			Usage struct {
+				CachedTokens *int `json:"cached_tokens"`
+			} `json:"usage"`
+		} `json:"choices"`
+	}
+
+	if err := common.Unmarshal(body, &payload); err != nil {
+		return 0, false
+	}
+
+	// 遍历choices查找cached_tokens
+	for _, choice := range payload.Choices {
+		if choice.Usage.CachedTokens != nil && *choice.Usage.CachedTokens > 0 {
+			return *choice.Usage.CachedTokens, true
+		}
+	}
+
+	return 0, false
+}
+
+// extractLlamaCachedTokensFromBody 从llama.cpp的非标准位置提取cache_n
+func extractLlamaCachedTokensFromBody(body []byte) (int, bool) {
+	if len(body) == 0 {
+		return 0, false
+	}
+
+	var payload struct {
+		Timings struct {
+			CachedTokens *int `json:"cache_n"`
+		} `json:"timings"`
+	}
+
+	if err := common.Unmarshal(body, &payload); err != nil {
+		return 0, false
+	}
+
+	if payload.Timings.CachedTokens == nil {
+		return 0, false
+	}
+	return *payload.Timings.CachedTokens, true
+}

relay/channel/tencent/relay-tencent.go

@@ -92,7 +92,7 @@ func streamResponseTencent2OpenAI(TencentResponse *TencentChatResponse) *dto.Cha
 
 func tencentStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
 	var responseText string
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 
 	helper.SetEventStreamHeaders(c)

relay/channel/vertex/adaptor.go

@@ -45,6 +45,7 @@ var claudeModelMap = map[string]string{
 	"claude-opus-4-5-20251101":   "claude-opus-4-5@20251101",
 	"claude-opus-4-6":            "claude-opus-4-6",
 	"claude-opus-4-7":            "claude-opus-4-7",
+	"claude-opus-4-8":            "claude-opus-4-8",
 }
 
 const anthropicVersion = "vertex-2023-10-16"

relay/channel/xai/adaptor.go

@@ -114,7 +114,7 @@ func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, request
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *types.NewAPIError) {
 	switch info.RelayMode {
 	case constant.RelayModeImagesGenerations, constant.RelayModeImagesEdits:
-		usage, err = openai.OpenaiHandlerWithUsage(c, info, resp)
+		usage, err = openai.OpenaiImageHandler(c, info, resp)
 	case constant.RelayModeResponses:
 		if info.IsStream {
 			usage, err = openai.OaiResponsesStreamHandler(c, info, resp)

relay/channel/zhipu/relay-zhipu.go

@@ -157,7 +157,7 @@ func streamMetaResponseZhipu2OpenAI(zhipuResponse *ZhipuStreamMetaResponse) (*dt
 
 func zhipuStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
 	var usage *dto.Usage
-	scanner := bufio.NewScanner(resp.Body)
+	scanner := helper.NewStreamScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 	dataChan := make(chan string)
 	metaChan := make(chan string)
@@ -180,6 +180,9 @@ func zhipuStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.
 				}
 			}
 		}
+		if err := scanner.Err(); err != nil {
+			common.SysLog("error reading stream: " + err.Error())
+		}
 		stopChan <- true
 	}()
 	helper.SetEventStreamHeaders(c)

relay/chat_completions_via_responses.go

@@ -1,7 +1,6 @@
 package relay
 
 import (
-	"bytes"
 	"io"
 	"net/http"
 	"strings"
@@ -125,7 +124,14 @@ func chatCompletionsViaResponses(c *gin.Context, info *relaycommon.RelayInfo, ad
 		return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
 	}
 
-	var requestBody io.Reader = bytes.NewBuffer(jsonData)
+	body, size, closer, err := relaycommon.NewOutboundJSONBody(jsonData)
+	if err != nil {
+		return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
+	}
+	defer closer.Close()
+	jsonData = nil
+	info.UpstreamRequestBodySize = size
+	var requestBody io.Reader = body
 
 	var httpResp *http.Response
 	resp, err := adaptor.DoRequest(c, info, requestBody)

relay/claude_handler.go

@@ -1,7 +1,6 @@
 package relay
 
 import (
-	"bytes"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -11,6 +10,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
@@ -53,14 +53,17 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 	}
 
 	if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(request.Model); ok && effortLevel != "" &&
-		(strings.HasPrefix(request.Model, "claude-opus-4-6") || strings.HasPrefix(request.Model, "claude-opus-4-7")) {
+		(strings.HasPrefix(request.Model, "claude-opus-4-6") ||
+			strings.HasPrefix(request.Model, "claude-opus-4-7") ||
+			strings.HasPrefix(request.Model, "claude-opus-4-8")) {
 		request.Model = baseModel
 		request.Thinking = &dto.Thinking{
 			Type: "adaptive",
 		}
 		request.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
-		if strings.HasPrefix(request.Model, "claude-opus-4-7") {
-			// Opus 4.7 rejects non-default temperature/top_p/top_k with 400
+		if strings.HasPrefix(request.Model, "claude-opus-4-7") ||
+			strings.HasPrefix(request.Model, "claude-opus-4-8") {
+			// Opus 4.7/4.8 reject non-default temperature/top_p/top_k with 400
 			// and defaults display to "omitted"; restore the 4.6 visible summary.
 			request.Thinking.Display = "summarized"
 			request.Temperature = nil
@@ -74,8 +77,9 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		strings.HasSuffix(request.Model, "-thinking") {
 		if request.Thinking == nil {
 			baseModel := strings.TrimSuffix(request.Model, "-thinking")
-			if strings.HasPrefix(baseModel, "claude-opus-4-7") {
-				// Opus 4.7 rejects thinking.type="enabled"; use adaptive at high effort.
+			if strings.HasPrefix(baseModel, "claude-opus-4-7") ||
+				strings.HasPrefix(baseModel, "claude-opus-4-8") {
+				// Opus 4.7/4.8 reject thinking.type="enabled"; use adaptive at high effort.
 				request.Thinking = &dto.Thinking{Type: "adaptive", Display: "summarized"}
 				request.OutputConfig = json.RawMessage(`{"effort":"high"}`)
 				request.Temperature = nil
@@ -151,6 +155,7 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		if err != nil {
 			return types.NewErrorWithStatusCode(err, types.ErrorCodeReadRequestBodyFailed, http.StatusBadRequest, types.ErrOptionWithSkipRetry())
 		}
+		info.UpstreamRequestBodySize = storage.Size()
 		requestBody = common.ReaderOnly(storage)
 	} else {
 		convertedRequest, err := adaptor.ConvertClaudeRequest(c, info, request)
@@ -177,10 +182,15 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 			}
 		}
 
-		if common.DebugEnabled {
-			println("requestBody: ", string(jsonData))
+		logger.LogDebug(c, "requestBody: %s", jsonData)
+		body, size, closer, err := relaycommon.NewOutboundJSONBody(jsonData)
+		if err != nil {
+			return types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
 		}
-		requestBody = bytes.NewBuffer(jsonData)
+		defer closer.Close()
+		jsonData = nil
+		info.UpstreamRequestBodySize = size
+		requestBody = body
 	}
 
 	statusCodeMappingStr := c.GetString("status_code_mapping")
@@ -202,7 +212,6 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 	}
 
 	usage, newAPIError := adaptor.DoResponse(c, httpResp, info)
-	//log.Printf("usage: %v", usage)
 	if newAPIError != nil {
 		// reset status code 重置状态码
 		service.ResetStatusCode(newAPIError, statusCodeMappingStr)

relay/common/outbound_body.go

@@ -0,0 +1,31 @@
+package common
+
+import (
+	"io"
+
+	"github.com/QuantumNous/new-api/common"
+)
+
+// NewOutboundJSONBody wraps the already-marshaled upstream request body into a
+// BodyStorage. When disk cache is enabled and the payload exceeds the configured
+// threshold, the data is written to a temp file and the original []byte can be
+// GC'd, significantly reducing the heap residency while waiting for the
+// upstream provider to respond (the dominant cost for large base64 payloads).
+//
+// In memory mode the underlying memoryStorage reuses the same backing array,
+// so this is equivalent to bytes.NewReader(data) in terms of memory usage.
+//
+// The caller MUST invoke closer.Close() once the upstream call has finished
+// (typically via defer) to release the disk file / memory accounting.
+//
+// The returned reader is wrapped with common.ReaderOnly to prevent the HTTP
+// transport from prematurely closing the underlying BodyStorage. The returned
+// size is meant to be propagated to http.Request.ContentLength because the
+// type-erased io.Reader prevents net/http from auto-detecting it.
+func NewOutboundJSONBody(data []byte) (body io.Reader, size int64, closer io.Closer, err error) {
+	storage, err := common.CreateBodyStorage(data)
+	if err != nil {
+		return nil, 0, nil, err
+	}
+	return common.ReaderOnly(storage), storage.Size(), storage, nil
+}

relay/common/override.go

@@ -26,13 +26,20 @@ const (
 
 var errSourceHeaderNotFound = errors.New("source header does not exist")
 
-var paramOverrideKeyAuditPaths = map[string]struct{}{
-	"model":          {},
-	"original_model": {},
-	"upstream_model": {},
-	"service_tier":   {},
-	"inference_geo":  {},
-	"speed":          {},
+var paramOverrideSensitivePathPrefixes = []string{
+	"model",
+	"original_model",
+	"upstream_model",
+	"service_tier",
+	"inference_geo",
+	"speed",
+	"messages",
+	"input",
+	"instructions",
+	"system",
+	"contents",
+	"systemInstruction",
+	"system_instruction",
 }
 
 type paramOverrideAuditRecorder struct {
@@ -146,9 +153,8 @@ func ApplyParamOverride(jsonData []byte, paramOverride map[string]interface{}, c
 			}
 		}
 
-		// 使用新方法
-		result, err := applyOperations(string(workingJSON), operations, conditionContext)
-		return []byte(result), err
+		// 使用新方法（基于 []byte，避免整包 string 拷贝）
+		return applyOperations(workingJSON, operations, conditionContext)
 	}
 
 	// 直接使用旧方法
@@ -206,6 +212,7 @@ func shouldEnableParamOverrideAudit(paramOverride map[string]interface{}) bool {
 	if operations, ok := tryParseOperations(paramOverride); ok {
 		for _, operation := range operations {
 			if shouldAuditParamPath(strings.TrimSpace(operation.Path)) ||
+				shouldAuditParamPath(strings.TrimSpace(operation.From)) ||
 				shouldAuditParamPath(strings.TrimSpace(operation.To)) {
 				return true
 			}
@@ -255,15 +262,19 @@ func shouldAuditParamPath(path string) bool {
 	if common.DebugEnabled {
 		return true
 	}
-	_, ok := paramOverrideKeyAuditPaths[path]
-	return ok
+	for _, prefix := range paramOverrideSensitivePathPrefixes {
+		if path == prefix || strings.HasPrefix(path, prefix+".") {
+			return true
+		}
+	}
+	return false
 }
 
 func shouldAuditOperation(mode, path, from, to string) bool {
 	if common.DebugEnabled {
 		return true
 	}
-	for _, candidate := range []string{path, to} {
+	for _, candidate := range []string{path, from, to} {
 		if shouldAuditParamPath(candidate) {
 			return true
 		}
@@ -498,13 +509,13 @@ func tryParseOperations(paramOverride map[string]interface{}) ([]ParamOperation,
 	return operations, true
 }
 
-func checkConditions(jsonStr, contextJSON string, conditions []ConditionOperation, logic string) (bool, error) {
+func checkConditions(data []byte, contextJSON string, conditions []ConditionOperation, logic string) (bool, error) {
 	if len(conditions) == 0 {
 		return true, nil // 没有条件，直接通过
 	}
 	results := make([]bool, len(conditions))
 	for i, condition := range conditions {
-		result, err := checkSingleCondition(jsonStr, contextJSON, condition)
+		result, err := checkSingleCondition(data, contextJSON, condition)
 		if err != nil {
 			return false, err
 		}
@@ -517,10 +528,10 @@ func checkConditions(jsonStr, contextJSON string, conditions []ConditionOperatio
 	return lo.SomeBy(results, func(item bool) bool { return item }), nil
 }
 
-func checkSingleCondition(jsonStr, contextJSON string, condition ConditionOperation) (bool, error) {
+func checkSingleCondition(data []byte, contextJSON string, condition ConditionOperation) (bool, error) {
 	// 处理负数索引
-	path := processNegativeIndex(jsonStr, condition.Path)
-	value := gjson.Get(jsonStr, path)
+	path := processNegativeIndex(data, condition.Path)
+	value := gjson.GetBytes(data, path)
 	if !value.Exists() && contextJSON != "" {
 		value = gjson.Get(contextJSON, condition.Path)
 	}
@@ -549,7 +560,7 @@ func checkSingleCondition(jsonStr, contextJSON string, condition ConditionOperat
 	return result, nil
 }
 
-func processNegativeIndex(jsonStr string, path string) string {
+func processNegativeIndex(data []byte, path string) string {
 	matches := negativeIndexRegexp.FindAllStringSubmatch(path, -1)
 
 	if len(matches) == 0 {
@@ -566,7 +577,7 @@ func processNegativeIndex(jsonStr string, path string) string {
 			arrayPath = arrayPath[:len(arrayPath)-1]
 		}
 
-		array := gjson.Get(jsonStr, arrayPath)
+		array := gjson.GetBytes(data, arrayPath)
 		if array.IsArray() {
 			length := len(array.Array())
 			actualIndex := length + index
@@ -655,36 +666,76 @@ func compareNumeric(jsonValue, targetValue gjson.Result, operator string) (bool,
 	}
 }
 
-// applyOperationsLegacy 原参数覆盖方法
+// applyOperationsLegacy 原参数覆盖方法。
+//
+// 旧实现把整个 jsonData unmarshal 成 map[string]interface{} 再 marshal 回来，
+// 对包含大 base64 字段（如 Gemini inlineData.data）的请求会放大数倍内存
+// （interface 装箱、map bucket、再次 marshal）。
+// 这里改成在 []byte 上直接调用 sjson.SetBytes，按顶层 key 逐个写入，
+// 不再把 payload 解码到 map[string]interface{}。
+//
+// 语义保持：每个 paramOverride 顶层 key 视为字面 key（不解析点号路径），
+// 与旧的 reqMap[key] = value 一致。包含 `.` `*` `?` `\` 的 key 会被转义，
+// 防止被 sjson 当作嵌套路径或通配符。
 func applyOperationsLegacy(jsonData []byte, paramOverride map[string]interface{}, auditRecorder *paramOverrideAuditRecorder) ([]byte, error) {
-	reqMap := make(map[string]interface{})
-	err := common.Unmarshal(jsonData, &reqMap)
-	if err != nil {
-		return nil, err
+	if len(paramOverride) == 0 {
+		return jsonData, nil
 	}
 
+	result := jsonData
 	for key, value := range paramOverride {
-		reqMap[key] = value
+		escaped := escapeSjsonLiteralKey(key)
+		next, err := sjson.SetBytes(result, escaped, value)
+		if err != nil {
+			return nil, err
+		}
+		result = next
 		auditRecorder.recordOperation("set", key, "", "", value)
 	}
 
-	return common.Marshal(reqMap)
+	return result, nil
 }
 
-func applyOperations(jsonStr string, operations []ParamOperation, conditionContext map[string]interface{}) (string, error) {
+// escapeSjsonLiteralKey 把可能被 sjson 误判为路径或通配符的字符转义，
+// 用于把字面 key 安全地传给 sjson.SetBytes / sjson.DeleteBytes。
+func escapeSjsonLiteralKey(key string) string {
+	if !strings.ContainsAny(key, ".*?\\") {
+		return key
+	}
+	var sb strings.Builder
+	sb.Grow(len(key) + 4)
+	for i := 0; i < len(key); i++ {
+		c := key[i]
+		switch c {
+		case '.', '*', '?', '\\':
+			sb.WriteByte('\\')
+		}
+		sb.WriteByte(c)
+	}
+	return sb.String()
+}
+
+// applyOperations 在 []byte 上原地应用所有 param override 操作。
+//
+// 旧实现走 string-based gjson/sjson，在 ApplyParamOverride 入口会做
+// string(jsonData) 与最终 []byte(result) 各一次整包拷贝，对大 base64
+// payload 来说每次重试都额外多花 2 倍 body 体积的临时内存。
+// 这里改成全程在 []byte 上工作，sjson.SetBytes / gjson.GetBytes 都是
+// 直接读写 []byte，每个操作只会产生一份新 buffer。
+func applyOperations(jsonData []byte, operations []ParamOperation, conditionContext map[string]interface{}) ([]byte, error) {
 	context := ensureContextMap(conditionContext)
 	auditRecorder := getParamOverrideAuditRecorder(context)
 	contextJSON, err := marshalContextJSON(context)
 	if err != nil {
-		return "", fmt.Errorf("failed to marshal condition context: %v", err)
+		return nil, fmt.Errorf("failed to marshal condition context: %v", err)
 	}
 
-	result := jsonStr
+	result := jsonData
 	for _, op := range operations {
 		// 检查条件是否满足
 		ok, err := checkConditions(result, contextJSON, op.Conditions, op.Logic)
 		if err != nil {
-			return "", err
+			return nil, err
 		}
 		if !ok {
 			continue // 条件不满足，跳过当前操作
@@ -695,7 +746,7 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 		if isPathBasedOperation(op.Mode) {
 			opPaths, err = resolveOperationPaths(result, opPath)
 			if err != nil {
-				return "", err
+				return nil, err
 			}
 			if len(opPaths) == 0 {
 				continue
@@ -713,10 +764,10 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 			}
 		case "set":
 			for _, path := range opPaths {
-				if op.KeepOrigin && gjson.Get(result, path).Exists() {
+				if op.KeepOrigin && gjson.GetBytes(result, path).Exists() {
 					continue
 				}
-				result, err = sjson.Set(result, path, op.Value)
+				result, err = sjson.SetBytes(result, path, op.Value)
 				if err != nil {
 					break
 				}
@@ -731,7 +782,7 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 			}
 		case "copy":
 			if op.From == "" || op.To == "" {
-				return "", fmt.Errorf("copy from/to is required")
+				return nil, fmt.Errorf("copy from/to is required")
 			}
 			opFrom := processNegativeIndex(result, op.From)
 			opTo := processNegativeIndex(result, op.To)
@@ -831,9 +882,9 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 			auditRecorder.recordOperation("return_error", op.Path, "", "", op.Value)
 			returnErr, parseErr := parseParamOverrideReturnError(op.Value)
 			if parseErr != nil {
-				return "", parseErr
+				return nil, parseErr
 			}
-			return "", returnErr
+			return nil, returnErr
 		case "prune_objects":
 			for _, path := range opPaths {
 				result, err = pruneObjects(result, path, contextJSON, op.Value)
@@ -890,7 +941,7 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 		case "pass_headers":
 			headerNames, parseErr := parseHeaderPassThroughNames(op.Value)
 			if parseErr != nil {
-				return "", parseErr
+				return nil, parseErr
 			}
 			for _, headerName := range headerNames {
 				if err = copyHeaderInContext(context, headerName, headerName, op.KeepOrigin); err != nil {
@@ -912,10 +963,10 @@ func applyOperations(jsonStr string, operations []ParamOperation, conditionConte
 				contextJSON, err = marshalContextJSON(context)
 			}
 		default:
-			return "", fmt.Errorf("unknown operation: %s", op.Mode)
+			return nil, fmt.Errorf("unknown operation: %s", op.Mode)
 		}
 		if err != nil {
-			return "", fmt.Errorf("operation %s failed: %w", op.Mode, err)
+			return nil, fmt.Errorf("operation %s failed: %w", op.Mode, err)
 		}
 	}
 	return result, nil
@@ -1349,11 +1400,11 @@ func parseSyncTarget(spec string) (syncTarget, error) {
 	}
 }
 
-func readSyncTargetValue(jsonStr string, context map[string]interface{}, target syncTarget) (interface{}, bool, error) {
+func readSyncTargetValue(data []byte, context map[string]interface{}, target syncTarget) (interface{}, bool, error) {
 	switch target.kind {
 	case "json":
-		path := processNegativeIndex(jsonStr, target.key)
-		value := gjson.Get(jsonStr, path)
+		path := processNegativeIndex(data, target.key)
+		value := gjson.GetBytes(data, path)
 		if !value.Exists() || value.Type == gjson.Null {
 			return nil, false, nil
 		}
@@ -1372,52 +1423,52 @@ func readSyncTargetValue(jsonStr string, context map[string]interface{}, target
 	}
 }
 
-func writeSyncTargetValue(jsonStr string, context map[string]interface{}, target syncTarget, value interface{}) (string, error) {
+func writeSyncTargetValue(data []byte, context map[string]interface{}, target syncTarget, value interface{}) ([]byte, error) {
 	switch target.kind {
 	case "json":
-		path := processNegativeIndex(jsonStr, target.key)
-		nextJSON, err := sjson.Set(jsonStr, path, value)
+		path := processNegativeIndex(data, target.key)
+		nextJSON, err := sjson.SetBytes(data, path, value)
 		if err != nil {
-			return "", err
+			return nil, err
 		}
 		return nextJSON, nil
 	case "header":
 		if err := setHeaderOverrideInContext(context, target.key, value, false); err != nil {
-			return "", err
+			return nil, err
 		}
-		return jsonStr, nil
+		return data, nil
 	default:
-		return "", fmt.Errorf("unsupported sync_fields target kind: %s", target.kind)
+		return nil, fmt.Errorf("unsupported sync_fields target kind: %s", target.kind)
 	}
 }
 
-func syncFieldsBetweenTargets(jsonStr string, context map[string]interface{}, fromSpec string, toSpec string) (string, error) {
+func syncFieldsBetweenTargets(data []byte, context map[string]interface{}, fromSpec string, toSpec string) ([]byte, error) {
 	fromTarget, err := parseSyncTarget(fromSpec)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 	toTarget, err := parseSyncTarget(toSpec)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 
-	fromValue, fromExists, err := readSyncTargetValue(jsonStr, context, fromTarget)
+	fromValue, fromExists, err := readSyncTargetValue(data, context, fromTarget)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
-	toValue, toExists, err := readSyncTargetValue(jsonStr, context, toTarget)
+	toValue, toExists, err := readSyncTargetValue(data, context, toTarget)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 
 	// If one side exists and the other side is missing, sync the missing side.
 	if fromExists && !toExists {
-		return writeSyncTargetValue(jsonStr, context, toTarget, fromValue)
+		return writeSyncTargetValue(data, context, toTarget, fromValue)
 	}
 	if toExists && !fromExists {
-		return writeSyncTargetValue(jsonStr, context, fromTarget, toValue)
+		return writeSyncTargetValue(data, context, fromTarget, toValue)
 	}
-	return jsonStr, nil
+	return data, nil
 }
 
 func ensureMapKeyInContext(context map[string]interface{}, key string) map[string]interface{} {
@@ -1491,24 +1542,24 @@ func syncRuntimeHeaderOverrideFromContext(info *RelayInfo, context map[string]in
 	info.UseRuntimeHeadersOverride = true
 }
 
-func moveValue(jsonStr, fromPath, toPath string) (string, error) {
-	sourceValue := gjson.Get(jsonStr, fromPath)
+func moveValue(data []byte, fromPath, toPath string) ([]byte, error) {
+	sourceValue := gjson.GetBytes(data, fromPath)
 	if !sourceValue.Exists() {
-		return jsonStr, fmt.Errorf("source path does not exist: %s", fromPath)
+		return data, fmt.Errorf("source path does not exist: %s", fromPath)
 	}
-	result, err := sjson.Set(jsonStr, toPath, sourceValue.Value())
+	result, err := sjson.SetBytes(data, toPath, sourceValue.Value())
 	if err != nil {
-		return "", err
+		return nil, err
 	}
-	return sjson.Delete(result, fromPath)
+	return sjson.DeleteBytes(result, fromPath)
 }
 
-func copyValue(jsonStr, fromPath, toPath string) (string, error) {
-	sourceValue := gjson.Get(jsonStr, fromPath)
+func copyValue(data []byte, fromPath, toPath string) ([]byte, error) {
+	sourceValue := gjson.GetBytes(data, fromPath)
 	if !sourceValue.Exists() {
-		return jsonStr, fmt.Errorf("source path does not exist: %s", fromPath)
+		return data, fmt.Errorf("source path does not exist: %s", fromPath)
 	}
-	return sjson.Set(jsonStr, toPath, sourceValue.Value())
+	return sjson.SetBytes(data, toPath, sourceValue.Value())
 }
 
 func isPathBasedOperation(mode string) bool {
@@ -1520,16 +1571,16 @@ func isPathBasedOperation(mode string) bool {
 	}
 }
 
-func resolveOperationPaths(jsonStr, path string) ([]string, error) {
+func resolveOperationPaths(data []byte, path string) ([]string, error) {
 	if !strings.Contains(path, "*") {
 		return []string{path}, nil
 	}
-	return expandWildcardPaths(jsonStr, path)
+	return expandWildcardPaths(data, path)
 }
 
-func expandWildcardPaths(jsonStr, path string) ([]string, error) {
+func expandWildcardPaths(data []byte, path string) ([]string, error) {
 	var root interface{}
-	if err := common.Unmarshal([]byte(jsonStr), &root); err != nil {
+	if err := common.Unmarshal(data, &root); err != nil {
 		return nil, err
 	}
 
@@ -1590,28 +1641,28 @@ func collectWildcardPaths(node interface{}, segments []string, prefix []string)
 	}
 }
 
-func deleteValue(jsonStr, path string) (string, error) {
+func deleteValue(data []byte, path string) ([]byte, error) {
 	if strings.TrimSpace(path) == "" {
-		return jsonStr, nil
+		return data, nil
 	}
-	return sjson.Delete(jsonStr, path)
+	return sjson.DeleteBytes(data, path)
 }
 
-func modifyValue(jsonStr, path string, value interface{}, keepOrigin, isPrepend bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func modifyValue(data []byte, path string, value interface{}, keepOrigin, isPrepend bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	switch {
 	case current.IsArray():
-		return modifyArray(jsonStr, path, value, isPrepend)
+		return modifyArray(data, path, value, isPrepend)
 	case current.Type == gjson.String:
-		return modifyString(jsonStr, path, value, isPrepend)
+		return modifyString(data, path, value, isPrepend)
 	case current.Type == gjson.JSON:
-		return mergeObjects(jsonStr, path, value, keepOrigin)
+		return mergeObjects(data, path, value, keepOrigin)
 	}
-	return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+	return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 }
 
-func modifyArray(jsonStr, path string, value interface{}, isPrepend bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func modifyArray(data []byte, path string, value interface{}, isPrepend bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	var newArray []interface{}
 	// 添加新值
 	addValue := func() {
@@ -1635,11 +1686,11 @@ func modifyArray(jsonStr, path string, value interface{}, isPrepend bool) (strin
 		addOriginal()
 		addValue()
 	}
-	return sjson.Set(jsonStr, path, newArray)
+	return sjson.SetBytes(data, path, newArray)
 }
 
-func modifyString(jsonStr, path string, value interface{}, isPrepend bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func modifyString(data []byte, path string, value interface{}, isPrepend bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	valueStr := fmt.Sprintf("%v", value)
 	var newStr string
 	if isPrepend {
@@ -1647,17 +1698,17 @@ func modifyString(jsonStr, path string, value interface{}, isPrepend bool) (stri
 	} else {
 		newStr = current.String() + valueStr
 	}
-	return sjson.Set(jsonStr, path, newStr)
+	return sjson.SetBytes(data, path, newStr)
 }
 
-func trimStringValue(jsonStr, path string, value interface{}, isPrefix bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func trimStringValue(data []byte, path string, value interface{}, isPrefix bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	if current.Type != gjson.String {
-		return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+		return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 	}
 
 	if value == nil {
-		return jsonStr, fmt.Errorf("trim value is required")
+		return data, fmt.Errorf("trim value is required")
 	}
 	valueStr := fmt.Sprintf("%v", value)
 
@@ -1667,69 +1718,69 @@ func trimStringValue(jsonStr, path string, value interface{}, isPrefix bool) (st
 	} else {
 		newStr = strings.TrimSuffix(current.String(), valueStr)
 	}
-	return sjson.Set(jsonStr, path, newStr)
+	return sjson.SetBytes(data, path, newStr)
 }
 
-func ensureStringAffix(jsonStr, path string, value interface{}, isPrefix bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func ensureStringAffix(data []byte, path string, value interface{}, isPrefix bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	if current.Type != gjson.String {
-		return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+		return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 	}
 
 	if value == nil {
-		return jsonStr, fmt.Errorf("ensure value is required")
+		return data, fmt.Errorf("ensure value is required")
 	}
 	valueStr := fmt.Sprintf("%v", value)
 	if valueStr == "" {
-		return jsonStr, fmt.Errorf("ensure value is required")
+		return data, fmt.Errorf("ensure value is required")
 	}
 
 	currentStr := current.String()
 	if isPrefix {
 		if strings.HasPrefix(currentStr, valueStr) {
-			return jsonStr, nil
+			return data, nil
 		}
-		return sjson.Set(jsonStr, path, valueStr+currentStr)
+		return sjson.SetBytes(data, path, valueStr+currentStr)
 	}
 
 	if strings.HasSuffix(currentStr, valueStr) {
-		return jsonStr, nil
+		return data, nil
 	}
-	return sjson.Set(jsonStr, path, currentStr+valueStr)
+	return sjson.SetBytes(data, path, currentStr+valueStr)
 }
 
-func transformStringValue(jsonStr, path string, transform func(string) string) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func transformStringValue(data []byte, path string, transform func(string) string) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	if current.Type != gjson.String {
-		return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+		return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 	}
-	return sjson.Set(jsonStr, path, transform(current.String()))
+	return sjson.SetBytes(data, path, transform(current.String()))
 }
 
-func replaceStringValue(jsonStr, path, from, to string) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func replaceStringValue(data []byte, path, from, to string) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	if current.Type != gjson.String {
-		return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+		return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 	}
 	if from == "" {
-		return jsonStr, fmt.Errorf("replace from is required")
+		return data, fmt.Errorf("replace from is required")
 	}
-	return sjson.Set(jsonStr, path, strings.ReplaceAll(current.String(), from, to))
+	return sjson.SetBytes(data, path, strings.ReplaceAll(current.String(), from, to))
 }
 
-func regexReplaceStringValue(jsonStr, path, pattern, replacement string) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func regexReplaceStringValue(data []byte, path, pattern, replacement string) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	if current.Type != gjson.String {
-		return jsonStr, fmt.Errorf("operation not supported for type: %v", current.Type)
+		return data, fmt.Errorf("operation not supported for type: %v", current.Type)
 	}
 	if pattern == "" {
-		return jsonStr, fmt.Errorf("regex pattern is required")
+		return data, fmt.Errorf("regex pattern is required")
 	}
 	re, err := regexp.Compile(pattern)
 	if err != nil {
-		return jsonStr, err
+		return data, err
 	}
-	return sjson.Set(jsonStr, path, re.ReplaceAllString(current.String(), replacement))
+	return sjson.SetBytes(data, path, re.ReplaceAllString(current.String(), replacement))
 }
 
 type pruneObjectsOptions struct {
@@ -1738,37 +1789,33 @@ type pruneObjectsOptions struct {
 	recursive  bool
 }
 
-func pruneObjects(jsonStr, path, contextJSON string, value interface{}) (string, error) {
+func pruneObjects(data []byte, path, contextJSON string, value interface{}) ([]byte, error) {
 	options, err := parsePruneObjectsOptions(value)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 
 	if path == "" {
 		var root interface{}
-		if err := common.Unmarshal([]byte(jsonStr), &root); err != nil {
-			return "", err
+		if err := common.Unmarshal(data, &root); err != nil {
+			return nil, err
 		}
 		cleaned, _, err := pruneObjectsNode(root, options, contextJSON, true)
 		if err != nil {
-			return "", err
+			return nil, err
 		}
-		cleanedBytes, err := common.Marshal(cleaned)
-		if err != nil {
-			return "", err
-		}
-		return string(cleanedBytes), nil
+		return common.Marshal(cleaned)
 	}
 
-	target := gjson.Get(jsonStr, path)
+	target := gjson.GetBytes(data, path)
 	if !target.Exists() {
-		return jsonStr, nil
+		return data, nil
 	}
 
 	var targetNode interface{}
 	if target.Type == gjson.JSON {
-		if err := common.Unmarshal([]byte(target.Raw), &targetNode); err != nil {
-			return "", err
+		if err := common.UnmarshalJsonStr(target.Raw, &targetNode); err != nil {
+			return nil, err
 		}
 	} else {
 		targetNode = target.Value()
@@ -1776,13 +1823,13 @@ func pruneObjects(jsonStr, path, contextJSON string, value interface{}) (string,
 
 	cleaned, _, err := pruneObjectsNode(targetNode, options, contextJSON, true)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 	cleanedBytes, err := common.Marshal(cleaned)
 	if err != nil {
-		return "", err
+		return nil, err
 	}
-	return sjson.SetRaw(jsonStr, path, string(cleanedBytes))
+	return sjson.SetRawBytes(data, path, cleanedBytes)
 }
 
 func parsePruneObjectsOptions(value interface{}) (pruneObjectsOptions, error) {
@@ -1958,16 +2005,16 @@ func shouldPruneObject(node map[string]interface{}, options pruneObjectsOptions,
 	if err != nil {
 		return false, err
 	}
-	return checkConditions(string(nodeBytes), contextJSON, options.conditions, options.logic)
+	return checkConditions(nodeBytes, contextJSON, options.conditions, options.logic)
 }
 
-func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (string, error) {
-	current := gjson.Get(jsonStr, path)
+func mergeObjects(data []byte, path string, value interface{}, keepOrigin bool) ([]byte, error) {
+	current := gjson.GetBytes(data, path)
 	var currentMap, newMap map[string]interface{}
 
-	// 解析当前值
-	if err := common.Unmarshal([]byte(current.Raw), &currentMap); err != nil {
-		return "", err
+	// 解析当前值（current.Raw 是 data 的子串，避免再分配一份）
+	if err := common.UnmarshalJsonStr(current.Raw, &currentMap); err != nil {
+		return nil, err
 	}
 	// 解析新值
 	switch v := value.(type) {
@@ -1976,7 +2023,7 @@ func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (str
 	default:
 		jsonBytes, _ := common.Marshal(v)
 		if err := common.Unmarshal(jsonBytes, &newMap); err != nil {
-			return "", err
+			return nil, err
 		}
 	}
 	// 合并
@@ -1989,7 +2036,7 @@ func mergeObjects(jsonStr, path string, value interface{}, keepOrigin bool) (str
 			result[k] = v
 		}
 	}
-	return sjson.Set(jsonStr, path, result)
+	return sjson.SetBytes(data, path, result)
 }
 
 // BuildParamOverrideContext 提供 ApplyParamOverride 可用的上下文信息。

relay/common/override_test.go

@@ -12,6 +12,7 @@ import (
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/setting/model_setting"
 	"github.com/samber/lo"
+	"github.com/stretchr/testify/require"
 )
 
 func TestApplyParamOverrideTrimPrefix(t *testing.T) {
@@ -2053,6 +2054,17 @@ func TestRemoveDisabledFieldsDefaultFiltering(t *testing.T) {
 	assertJSONEqual(t, `{"cache_control":{"type":"ephemeral"},"store":true}`, string(out))
 }
 
+func TestRemoveDisabledFieldsNoControlledFieldsKeepsBody(t *testing.T) {
+	input := `{"model":"gpt-4o","messages":[{"role":"user","content":"hi"}]}`
+	settings := dto.ChannelOtherSettings{}
+
+	out, err := RemoveDisabledFields([]byte(input), settings, false)
+	if err != nil {
+		t.Fatalf("RemoveDisabledFields returned error: %v", err)
+	}
+	require.Equal(t, input, string(out))
+}
+
 func TestRemoveDisabledFieldsAllowInferenceGeo(t *testing.T) {
 	input := `{
 		"inference_geo":"eu",
@@ -2184,6 +2196,95 @@ func TestApplyParamOverrideWithRelayInfoRecordsOnlyKeyOperationsWhenDebugDisable
 	}
 }
 
+func TestApplyParamOverrideWithRelayInfoRecordsConversationBodyOperationsWhenDebugDisabled(t *testing.T) {
+	originalDebugEnabled := common2.DebugEnabled
+	common2.DebugEnabled = false
+	t.Cleanup(func() {
+		common2.DebugEnabled = originalDebugEnabled
+	})
+
+	info := &RelayInfo{
+		ChannelMeta: &ChannelMeta{
+			ParamOverride: map[string]interface{}{
+				"operations": []interface{}{
+					map[string]interface{}{
+						"mode": "replace",
+						"path": "messages.0.content",
+						"from": "hello",
+						"to":   "hi",
+					},
+					map[string]interface{}{
+						"mode":  "set",
+						"path":  "input.0.content.0.text",
+						"value": "rewritten response input",
+					},
+					map[string]interface{}{
+						"mode":  "set",
+						"path":  "instructions",
+						"value": "new instruction",
+					},
+					map[string]interface{}{
+						"mode":  "append",
+						"path":  "contents.0.parts",
+						"value": map[string]interface{}{"text": "new gemini part"},
+					},
+					map[string]interface{}{
+						"mode": "copy",
+						"from": "system",
+						"to":   "metadata.system_copy",
+					},
+					map[string]interface{}{
+						"mode":  "set",
+						"path":  "temperature",
+						"value": 0.1,
+					},
+				},
+			},
+		},
+	}
+
+	out, err := ApplyParamOverrideWithRelayInfo([]byte(`{
+		"messages":[{"role":"user","content":"hello world"}],
+		"input":[{"role":"user","content":[{"type":"input_text","text":"original response input"}]}],
+		"instructions":"old instruction",
+		"system":"old system",
+		"contents":[{"role":"user","parts":[{"text":"hello gemini"}]}],
+		"temperature":0.7
+	}`), info)
+	require.NoError(t, err)
+	assertJSONEqual(t, `{
+		"messages":[{"role":"user","content":"hi world"}],
+		"input":[{"role":"user","content":[{"type":"input_text","text":"rewritten response input"}]}],
+		"instructions":"new instruction",
+		"system":"old system",
+		"contents":[{"role":"user","parts":[{"text":"hello gemini"},{"text":"new gemini part"}]}],
+		"temperature":0.1,
+		"metadata":{"system_copy":"old system"}
+	}`, string(out))
+
+	require.Equal(t, []string{
+		"replace messages.0.content from hello to hi",
+		"set input.0.content.0.text = rewritten response input",
+		"set instructions = new instruction",
+		"append contents.0.parts with {\"text\":\"new gemini part\"}",
+		"copy system -> metadata.system_copy",
+	}, info.ParamOverrideAudit)
+}
+
+func TestShouldAuditParamPathUsesFieldBoundaryPrefixMatching(t *testing.T) {
+	originalDebugEnabled := common2.DebugEnabled
+	common2.DebugEnabled = false
+	t.Cleanup(func() {
+		common2.DebugEnabled = originalDebugEnabled
+	})
+
+	require.True(t, shouldAuditParamPath("messages"))
+	require.True(t, shouldAuditParamPath("messages.0.content"))
+	require.True(t, shouldAuditParamPath("systemInstruction.parts.0.text"))
+	require.False(t, shouldAuditParamPath("model_name"))
+	require.False(t, shouldAuditParamPath("message"))
+}
+
 func assertJSONEqual(t *testing.T, want, got string) {
 	t.Helper()
 

relay/common/relay_info.go

@@ -18,6 +18,7 @@ import (
 
 	"github.com/gin-gonic/gin"
 	"github.com/gorilla/websocket"
+	"github.com/tidwall/gjson"
 )
 
 type ThinkingContentInfo struct {
@@ -153,6 +154,13 @@ type RelayInfo struct {
 	UseRuntimeHeadersOverride             bool
 	ParamOverrideAudit                    []string
 
+	// UpstreamRequestBodySize is the byte size of the marshaled upstream request
+	// body. It is set when the body is wrapped in a BodyStorage (see
+	// relay/common/outbound_body.go), so that DoApiRequest can populate
+	// http.Request.ContentLength manually (net/http only auto-detects it for
+	// *bytes.Reader/Buffer/strings.Reader). 0 means "let net/http decide".
+	UpstreamRequestBodySize int64
+
 	PriceData types.PriceData
 
 	// TieredBillingSnapshot is a frozen snapshot of tiered billing rules
@@ -785,6 +793,9 @@ func RemoveDisabledFields(jsonData []byte, channelOtherSettings dto.ChannelOther
 	if model_setting.GetGlobalSettings().PassThroughRequestEnabled || channelPassThroughEnabled {
 		return jsonData, nil
 	}
+	if !hasRemovableDisabledField(jsonData, channelOtherSettings) {
+		return jsonData, nil
+	}
 
 	var data map[string]interface{}
 	if err := common.Unmarshal(jsonData, &data); err != nil {
@@ -851,6 +862,25 @@ func RemoveDisabledFields(jsonData []byte, channelOtherSettings dto.ChannelOther
 	return jsonDataAfter, nil
 }
 
+func hasRemovableDisabledField(jsonData []byte, channelOtherSettings dto.ChannelOtherSettings) bool {
+	values := gjson.GetManyBytes(
+		jsonData,
+		"service_tier",
+		"inference_geo",
+		"speed",
+		"store",
+		"safety_identifier",
+		"stream_options.include_obfuscation",
+	)
+
+	return (!channelOtherSettings.AllowServiceTier && values[0].Exists()) ||
+		(!channelOtherSettings.AllowInferenceGeo && values[1].Exists()) ||
+		(!channelOtherSettings.AllowSpeed && values[2].Exists()) ||
+		(channelOtherSettings.DisableStore && values[3].Exists()) ||
+		(!channelOtherSettings.AllowSafetyIdentifier && values[4].Exists()) ||
+		(!channelOtherSettings.AllowIncludeObfuscation && values[5].Exists())
+}
+
 // RemoveGeminiDisabledFields removes disabled fields from Gemini request JSON data
 // Currently supports removing functionResponse.id field which Vertex AI does not support
 func RemoveGeminiDisabledFields(jsonData []byte) ([]byte, error) {

relay/common_handler/rerank.go

@@ -7,6 +7,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/relay/channel/xinference"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/service"
@@ -21,9 +22,7 @@ func RerankHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
 	}
 	service.CloseResponseBodyGracefully(resp)
-	if common.DebugEnabled {
-		println("reranker response body: ", string(responseBody))
-	}
+	logger.LogDebug(c, "reranker response body: %s", responseBody)
 	var jinaResp dto.RerankResponse
 	if info.ChannelType == constant.ChannelTypeXinference {
 		var xinRerankResponse xinference.XinRerankResponse

relay/compatible_handler.go

@@ -1,7 +1,6 @@
 package relay
 
 import (
-	"bytes"
 	"fmt"
 	"io"
 	"net/http"
@@ -102,7 +101,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 		}
 		if common.DebugEnabled {
 			if debugBytes, bErr := storage.Bytes(); bErr == nil {
-				println("requestBody: ", string(debugBytes))
+				logger.LogDebug(c, "requestBody: %s", debugBytes)
 			}
 		}
 		requestBody = common.ReaderOnly(storage)
@@ -174,9 +173,16 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 			}
 		}
 
-		logger.LogDebug(c, fmt.Sprintf("text request body: %s", string(jsonData)))
+		logger.LogDebug(c, "text request body: %s", jsonData)
 
-		requestBody = bytes.NewBuffer(jsonData)
+		body, size, closer, err := relaycommon.NewOutboundJSONBody(jsonData)
+		if err != nil {
+			return types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
+		}
+		defer closer.Close()
+		jsonData = nil
+		info.UpstreamRequestBodySize = size
+		requestBody = body
 	}
 
 	var httpResp *http.Response

relay/embedding_handler.go

@@ -1,7 +1,6 @@
 package relay
 
 import (
-	"bytes"
 	"fmt"
 	"io"
 	"net/http"
@@ -58,8 +57,15 @@ func EmbeddingHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 		}
 	}
 
-	logger.LogDebug(c, fmt.Sprintf("converted embedding request body: %s", string(jsonData)))
-	var requestBody io.Reader = bytes.NewBuffer(jsonData)
+	logger.LogDebug(c, "converted embedding request body: %s", jsonData)
+	body, size, closer, err := relaycommon.NewOutboundJSONBody(jsonData)
+	if err != nil {
+		return types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
+	}
+	defer closer.Close()
+	jsonData = nil
+	info.UpstreamRequestBodySize = size
+	var requestBody io.Reader = body
 	statusCodeMappingStr := c.GetString("status_code_mapping")
 	resp, err := adaptor.DoRequest(c, info, requestBody)
 	if err != nil {