Merge pull request #4076 from seefs001/ci/add-pr-check

ci: refine PR template and add PR submission checks

.env.example

@@ -19,6 +19,8 @@
 # HOSTNAME=your-hostname
 
 # 数据库相关配置
+# 启用错误日志记录
+# ERROR_LOG_ENABLED=true
 # 数据库连接字符串
 # SQL_DSN=user:password@tcp(127.0.0.1:3306)/dbname?parseTime=true
 # 日志数据库连接字符串

.github/FUNDING.yml

@@ -1,12 +0,0 @@
-# These are supported funding model platforms
-
-github: # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
-patreon: # Replace with a single Patreon username
-open_collective: # Replace with a single Open Collective username
-ko_fi: # Replace with a single Ko-fi username
-tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
-community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
-liberapay: # Replace with a single Liberapay username
-issuehunt: # Replace with a single IssueHunt username
-otechie: # Replace with a single Otechie username
-custom: ['https://afdian.com/a/new-api'] # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,28 @@
+# ⚠️ 提交说明 / PR Notice
+> [!IMPORTANT]
+>
+> - 请提供**人工撰写**的简洁摘要，避免直接粘贴未经整理的 AI 输出。
+
+## 📝 变更描述 / Description
+(简述：做了什么？为什么这样改能生效？请基于你对代码逻辑的理解来写，避免粘贴未经整理的内容)
+
+## 🚀 变更类型 / Type of change
+- [ ] 🐛 Bug 修复 (Bug fix) - *请关联对应 Issue，避免将设计取舍、理解偏差或预期不一致直接归类为 bug*
+- [ ] ✨ 新功能 (New feature) - *重大特性建议先通过 Issue 沟通*
+- [ ] ⚡ 性能优化 / 重构 (Refactor)
+- [ ] 📝 文档更新 (Documentation)
+
+## 🔗 关联任务 / Related Issue
+- Closes # (如有)
+
+## ✅ 提交前检查项 / Checklist
+- [ ] **人工确认:** 我已亲自整理并撰写此描述，没有直接粘贴未经处理的 AI 输出。
+- [ ] **非重复提交:** 我已搜索现有的 [Issues](https://github.com/QuantumNous/new-api/issues) 与 [PRs](https://github.com/QuantumNous/new-api/pulls)，确认不是重复提交。
+- [ ] **Bug fix 说明:** 若此 PR 标记为 `Bug fix`，我已提交或关联对应 Issue，且不会将设计取舍、预期不一致或理解偏差直接归类为 bug。
+- [ ] **变更理解:** 我已理解这些更改的工作原理及可能影响。
+- [ ] **范围聚焦:** 本 PR 未包含任何与当前任务无关的代码改动。
+- [ ] **本地验证:** 已在本地运行并通过测试或手动验证，维护者可以据此复核结果。
+- [ ] **安全合规:** 代码中无敏感凭据，且符合项目代码规范。
+
+## 📸 运行证明 / Proof of Work
+(请在此粘贴截图、关键日志或测试报告，以证明变更生效)

.github/PULL_REQUEST_TEMPLATE/pull_request_template.md

@@ -1,29 +0,0 @@
-# ⚠️ 提交警告 / PR Warning
-> **请注意：** 请提供**人工撰写**的简洁摘要。包含大量 AI 灌水内容、逻辑混乱或无视模版的 PR **可能会被无视或直接关闭**。
-
----
-
-## 💡 沟通提示 / Pre-submission
-> **重大功能变更？** 请先提交 Issue 交流，避免无效劳动。
-
-## 📝 变更描述 / Description
-(简述：做了什么？为什么这样改能生效？你必须理解代码逻辑，禁止粘贴 AI 废话)
-
-## 🚀 变更类型 / Type of change
-- [ ] 🐛 Bug 修复 (Bug fix)
-- [ ] ✨ 新功能 (New feature) - *重大特性建议先 Issue 沟通*
-- [ ] ⚡ 性能优化 / 重构 (Refactor)
-- [ ] 📝 文档更新 (Documentation)
-
-## 🔗 关联任务 / Related Issue
-- Closes # (如有)
-
-## ✅ 提交前检查项 / Checklist
-- [ ] **人工确认:** 我已亲自撰写此描述，去除了 AI 原始输出的冗余。
-- [ ] **深度理解:** 我已**完全理解**这些更改的工作原理及潜在影响。
-- [ ] **范围聚焦:** 本 PR 未包含任何与当前任务无关的代码改动。
-- [ ] **本地验证:** 已在本地运行并通过了测试或手动验证。
-- [ ] **安全合规:** 代码中无敏感凭据，且符合项目代码规范。
-
-## 📸 运行证明 / Proof of Work
-(请在此粘贴截图、关键日志或测试报告，以证明变更生效)

.github/workflows/docker-image-alpha.yml

@@ -27,9 +27,10 @@ jobs:
     permissions:
       packages: write
       contents: read
+      id-token: write
     steps:
       - name: Check out (shallow)
-        uses: actions/checkout@v4
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: 1
 
@@ -46,16 +47,16 @@ jobs:
         run: echo "GHCR_REPOSITORY=${GITHUB_REPOSITORY,,}" >> $GITHUB_ENV
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f # v3
 
       - name: Log in to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
       - name: Log in to GHCR
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           registry: ghcr.io
           username: ${{ github.actor }}
@@ -63,14 +64,15 @@ jobs:
 
       - name: Extract metadata (labels)
         id: meta
-        uses: docker/metadata-action@v5
+        uses: docker/metadata-action@c299e40c65443455700f0fdfc63efafe5b349051 # v5
         with:
           images: |
             calciumion/new-api
             ghcr.io/${{ env.GHCR_REPOSITORY }}
 
       - name: Build & push single-arch (to both registries)
-        uses: docker/build-push-action@v6
+        id: build
+        uses: docker/build-push-action@10e90e3645eae34f1e60eeb005ba3a3d33f178e8 # v6
         with:
           context: .
           platforms: ${{ matrix.platform }}
@@ -83,8 +85,25 @@ jobs:
           labels: ${{ steps.meta.outputs.labels }}
           cache-from: type=gha
           cache-to: type=gha,mode=max
-          provenance: false
-          sbom: false
+          provenance: mode=max
+          sbom: true
+
+      - name: Install cosign
+        uses: sigstore/cosign-installer@398d4b0eeef1380460a10c8013a76f728fb906ac # v3
+
+      - name: Sign image with cosign
+        run: |
+          cosign sign --yes calciumion/new-api@${{ steps.build.outputs.digest }}
+          cosign sign --yes ghcr.io/${{ env.GHCR_REPOSITORY }}@${{ steps.build.outputs.digest }}
+
+      - name: Output digest
+        run: |
+          echo "### Docker Image Digest (${{ matrix.arch }})" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          echo "calciumion/new-api:alpha-${{ matrix.arch }}" >> $GITHUB_STEP_SUMMARY
+          echo "ghcr.io/${{ env.GHCR_REPOSITORY }}:alpha-${{ matrix.arch }}" >> $GITHUB_STEP_SUMMARY
+          echo "${{ steps.build.outputs.digest }}" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
 
   create_manifests:
     name: Create multi-arch manifests (Docker Hub + GHCR)
@@ -95,7 +114,7 @@ jobs:
       contents: read
     steps:
       - name: Check out (shallow)
-        uses: actions/checkout@v4
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: 1
 
@@ -110,7 +129,7 @@ jobs:
           echo "VERSION=$VERSION" >> $GITHUB_ENV
 
       - name: Log in to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
@@ -130,7 +149,7 @@ jobs:
             calciumion/new-api:${VERSION}-arm64
 
       - name: Log in to GHCR
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           registry: ghcr.io
           username: ${{ github.actor }}
@@ -149,3 +168,12 @@ jobs:
             -t ghcr.io/${GHCR_REPOSITORY}:${VERSION} \
             ghcr.io/${GHCR_REPOSITORY}:${VERSION}-amd64 \
             ghcr.io/${GHCR_REPOSITORY}:${VERSION}-arm64
+
+      - name: Output manifest digest
+        run: |
+          echo "### Multi-arch Manifest Digests" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          docker buildx imagetools inspect calciumion/new-api:alpha >> $GITHUB_STEP_SUMMARY
+          echo "---" >> $GITHUB_STEP_SUMMARY
+          docker buildx imagetools inspect ghcr.io/${GHCR_REPOSITORY}:alpha >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY

.github/workflows/docker-image-arm64.yml

@@ -30,10 +30,11 @@ jobs:
     permissions:
       packages: write
       contents: read
+      id-token: write
 
     steps:
       - name: Check out
-        uses: actions/checkout@v4
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: ${{ github.event_name == 'workflow_dispatch' && 0 || 1 }}
           ref: ${{ github.event.inputs.tag || github.ref }}
@@ -59,16 +60,16 @@ jobs:
 #        run: echo "GHCR_REPOSITORY=${GITHUB_REPOSITORY,,}" >> $GITHUB_ENV
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f # v3
 
       - name: Log in to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
 #      - name: Log in to GHCR
-#        uses: docker/login-action@v3
+#        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
 #        with:
 #          registry: ghcr.io
 #          username: ${{ github.actor }}
@@ -76,14 +77,15 @@ jobs:
 
       - name: Extract metadata (labels)
         id: meta
-        uses: docker/metadata-action@v5
+        uses: docker/metadata-action@c299e40c65443455700f0fdfc63efafe5b349051 # v5
         with:
           images: |
             calciumion/new-api
 #            ghcr.io/${{ env.GHCR_REPOSITORY }}
 
       - name: Build & push single-arch (to both registries)
-        uses: docker/build-push-action@v6
+        id: build
+        uses: docker/build-push-action@10e90e3645eae34f1e60eeb005ba3a3d33f178e8 # v6
         with:
           context: .
           platforms: ${{ matrix.platform }}
@@ -96,8 +98,22 @@ jobs:
           labels: ${{ steps.meta.outputs.labels }}
           cache-from: type=gha
           cache-to: type=gha,mode=max
-          provenance: false
-          sbom: false
+          provenance: mode=max
+          sbom: true
+
+      - name: Install cosign
+        uses: sigstore/cosign-installer@398d4b0eeef1380460a10c8013a76f728fb906ac # v3
+
+      - name: Sign image with cosign
+        run: cosign sign --yes calciumion/new-api@${{ steps.build.outputs.digest }}
+
+      - name: Output digest
+        run: |
+          echo "### Docker Image Digest (${{ matrix.arch }})" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          echo "calciumion/new-api:${{ env.TAG }}-${{ matrix.arch }}" >> $GITHUB_STEP_SUMMARY
+          echo "${{ steps.build.outputs.digest }}" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
 
   create_manifests:
     name: Create multi-arch manifests (Docker Hub)
@@ -117,7 +133,7 @@ jobs:
 #        run: echo "GHCR_REPOSITORY=${GITHUB_REPOSITORY,,}" >> $GITHUB_ENV
 
       - name: Log in to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
@@ -136,9 +152,16 @@ jobs:
             calciumion/new-api:latest-amd64 \
             calciumion/new-api:latest-arm64
 
+      - name: Output manifest digest
+        run: |
+          echo "### Multi-arch Manifest" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          docker buildx imagetools inspect calciumion/new-api:${TAG} >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+
       # ---- GHCR ----
 #      - name: Log in to GHCR
-#        uses: docker/login-action@v3
+#        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # v3
 #        with:
 #          registry: ghcr.io
 #          username: ${{ github.actor }}

.github/workflows/pr-check.yml

@@ -0,0 +1,33 @@
+name: PR Check
+
+permissions:
+  contents: read
+  issues: read
+  pull-requests: read
+
+on:
+  pull_request_target:
+    types: [opened, reopened]
+
+jobs:
+  pr-quality:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: peakoss/anti-slop@v0.2.1
+        with:
+          max-failures: 4
+          require-description: true
+
+          # require-linked-issue: false
+          blocked-terms:  |
+            🤖 Generated with Claude Code
+
+          require-pr-template: true
+          strict-pr-template-sections: "✅ 提交前检查项 / Checklist"
+
+          detect-spam-usernames: true
+          min-account-age: 30
+
+          failure-add-pr-labels: "pr-check-failed"
+          failure-pr-message: "感谢您的提交。由于该 PR 未遵循我们的贡献模板，且被识别为缺乏人工参与的纯 AI 生成内容 (AI Slop)，我们将先予以关闭。我们更欢迎经过人工审核、验证并带有个人思考的贡献。如果您认为这其中存在误解，请回复告知。/ Thank you for your submission. This PR has been closed because it does not follow our contribution template and has been identified as purely AI-generated content (AI Slop) without meaningful human involvement. We prioritize contributions that are human-verified and reflect individual effort. If you believe this is a mistake, please let us know by replying to this comment."
+          close-pr: true

.github/workflows/release.yml

@@ -19,14 +19,14 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: 0
       - name: Determine Version
         run: |
           VERSION=$(git describe --tags)
           echo "VERSION=$VERSION" >> $GITHUB_ENV
-      - uses: oven-sh/setup-bun@v2
+      - uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2
         with:
           bun-version: latest
       - name: Build Frontend
@@ -38,7 +38,7 @@ jobs:
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
-        uses: actions/setup-go@v3
+        uses: actions/setup-go@40f1582b2485089dde7abd97c1529aa768e1baff # v5
         with:
           go-version: '>=1.25.1'
       - name: Build Backend (amd64)
@@ -50,12 +50,16 @@ jobs:
           sudo apt-get update
           DEBIAN_FRONTEND=noninteractive sudo apt-get install -y gcc-aarch64-linux-gnu
           CC=aarch64-linux-gnu-gcc CGO_ENABLED=1 GOOS=linux GOARCH=arm64 go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION' -extldflags '-static'" -o new-api-arm64-$VERSION
+      - name: Generate checksums
+        run: sha256sum new-api-* > checksums-linux.txt
+
       - name: Release
-        uses: softprops/action-gh-release@v2
+        uses: softprops/action-gh-release@153bb8e04406b158c6c84fc1615b65b24149a1fe # v2
         if: startsWith(github.ref, 'refs/tags/')
         with:
           files: |
             new-api-*
+            checksums-linux.txt
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
@@ -64,14 +68,14 @@ jobs:
     runs-on: macos-latest
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: 0
       - name: Determine Version
         run: |
           VERSION=$(git describe --tags)
           echo "VERSION=$VERSION" >> $GITHUB_ENV
-      - uses: oven-sh/setup-bun@v2
+      - uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2
         with:
           bun-version: latest
       - name: Build Frontend
@@ -84,18 +88,23 @@ jobs:
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
-        uses: actions/setup-go@v3
+        uses: actions/setup-go@40f1582b2485089dde7abd97c1529aa768e1baff # v5
         with:
           go-version: '>=1.25.1'
       - name: Build Backend
         run: |
           go mod download
           go build -ldflags "-X 'new-api/common.Version=$VERSION'" -o new-api-macos-$VERSION
+      - name: Generate checksums
+        run: shasum -a 256 new-api-macos-* > checksums-macos.txt
+
       - name: Release
-        uses: softprops/action-gh-release@v2
+        uses: softprops/action-gh-release@153bb8e04406b158c6c84fc1615b65b24149a1fe # v2
         if: startsWith(github.ref, 'refs/tags/')
         with:
-          files: new-api-macos-*
+          files: |
+            new-api-macos-*
+            checksums-macos.txt
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
@@ -107,14 +116,14 @@ jobs:
         shell: bash
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
         with:
           fetch-depth: 0
       - name: Determine Version
         run: |
           VERSION=$(git describe --tags)
           echo "VERSION=$VERSION" >> $GITHUB_ENV
-      - uses: oven-sh/setup-bun@v2
+      - uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2
         with:
           bun-version: latest
       - name: Build Frontend
@@ -126,17 +135,22 @@ jobs:
           DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
           cd ..
       - name: Set up Go
-        uses: actions/setup-go@v3
+        uses: actions/setup-go@40f1582b2485089dde7abd97c1529aa768e1baff # v5
         with:
           go-version: '>=1.25.1'
       - name: Build Backend
         run: |
           go mod download
           go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION'" -o new-api-$VERSION.exe
+      - name: Generate checksums
+        run: sha256sum new-api-*.exe > checksums-windows.txt
+
       - name: Release
-        uses: softprops/action-gh-release@v2
+        uses: softprops/action-gh-release@153bb8e04406b158c6c84fc1615b65b24149a1fe # v2
         if: startsWith(github.ref, 'refs/tags/')
         with:
-          files: new-api-*.exe
+          files: |
+            new-api-*.exe
+            checksums-windows.txt
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

Dockerfile

@@ -1,4 +1,4 @@
-FROM oven/bun:latest AS builder
+FROM oven/bun:1@sha256:0733e50325078969732ebe3b15ce4c4be5082f18c4ac1a0f0ca4839c2e4e42a7 AS builder
 
 WORKDIR /build
 COPY web/package.json .
@@ -8,7 +8,7 @@ COPY ./web .
 COPY ./VERSION .
 RUN DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
 
-FROM golang:alpine AS builder2
+FROM golang:1.26.1-alpine@sha256:2389ebfa5b7f43eeafbd6be0c3700cc46690ef842ad962f6c5bd6be49ed82039 AS builder2
 ENV GO111MODULE=on CGO_ENABLED=0
 
 ARG TARGETOS
@@ -25,7 +25,7 @@ COPY . .
 COPY --from=builder /build/dist ./web/dist
 RUN go build -ldflags "-s -w -X 'github.com/QuantumNous/new-api/common.Version=$(cat VERSION)'" -o new-api
 
-FROM debian:bookworm-slim
+FROM debian:bookworm-slim@sha256:f06537653ac770703bc45b4b113475bd402f451e85223f0f2837acbf89ab020a
 
 RUN apt-get update \
     && apt-get install -y --no-install-recommends ca-certificates tzdata libasan8 wget \

README.zh_CN.md

@@ -383,7 +383,7 @@ docker run --name new-api -d --restart always \
 2. 在应用商店搜索 **New-API**
 3. 一键安装
 
-📖 [图文教程](./docs/BT.md)
+📖 [图文教程](./docs/installation/BT.md)
 
 </details>
 

README.zh_TW.md

@@ -70,17 +70,20 @@
 <p align="center">
   <a href="https://www.cherry-ai.com/" target="_blank">
     <img src="./docs/images/cherry-studio.png" alt="Cherry Studio" height="80" />
-  </a>
-  <a href="https://bda.pku.edu.cn/" target="_blank">
+  </a><!--
+  --><a href="https://github.com/iOfficeAI/AionUi/" target="_blank">
+    <img src="./docs/images/aionui.png" alt="Aion UI" height="80" />
+  </a><!--
+  --><a href="https://bda.pku.edu.cn/" target="_blank">
     <img src="./docs/images/pku.png" alt="北京大學" height="80" />
-  </a>
-  <a href="https://www.compshare.cn/?ytag=GPU_yy_gh_newapi" target="_blank">
+  </a><!--
+  --><a href="https://www.compshare.cn/?ytag=GPU_yy_gh_newapi" target="_blank">
     <img src="./docs/images/ucloud.png" alt="UCloud 優刻得" height="80" />
-  </a>
-  <a href="https://www.aliyun.com/" target="_blank">
+  </a><!--
+  --><a href="https://www.aliyun.com/" target="_blank">
     <img src="./docs/images/aliyun.png" alt="阿里雲" height="80" />
-  </a>
-  <a href="https://io.net/" target="_blank">
+  </a><!--
+  --><a href="https://io.net/" target="_blank">
     <img src="./docs/images/io-net.png" alt="IO.NET" height="80" />
   </a>
 </p>

common/constants.go

@@ -80,6 +80,7 @@ var InsecureTLSConfig = &tls.Config{InsecureSkipVerify: true}
 var SMTPServer = ""
 var SMTPPort = 587
 var SMTPSSLEnabled = false
+var SMTPForceAuthLogin = false
 var SMTPAccount = ""
 var SMTPFrom = ""
 var SMTPToken = ""

common/email.go

@@ -19,6 +19,20 @@ func generateMessageID() (string, error) {
 	return fmt.Sprintf("<%d.%s@%s>", time.Now().UnixNano(), GetRandomString(12), domain), nil
 }
 
+func shouldUseSMTPLoginAuth() bool {
+	if SMTPForceAuthLogin {
+		return true
+	}
+	return isOutlookServer(SMTPAccount) || slices.Contains(EmailLoginAuthServerList, SMTPServer)
+}
+
+func getSMTPAuth() smtp.Auth {
+	if shouldUseSMTPLoginAuth() {
+		return LoginAuth(SMTPAccount, SMTPToken)
+	}
+	return smtp.PlainAuth("", SMTPAccount, SMTPToken, SMTPServer)
+}
+
 func SendEmail(subject string, receiver string, content string) error {
 	if SMTPFrom == "" { // for compatibility
 		SMTPFrom = SMTPAccount
@@ -38,7 +52,7 @@ func SendEmail(subject string, receiver string, content string) error {
 		"Message-ID: %s\r\n"+ // 添加 Message-ID 头
 		"Content-Type: text/html; charset=UTF-8\r\n\r\n%s\r\n",
 		receiver, SystemName, SMTPFrom, encodedSubject, time.Now().Format(time.RFC1123Z), id, content))
-	auth := smtp.PlainAuth("", SMTPAccount, SMTPToken, SMTPServer)
+	auth := getSMTPAuth()
 	addr := fmt.Sprintf("%s:%d", SMTPServer, SMTPPort)
 	to := strings.Split(receiver, ";")
 	var err error
@@ -80,9 +94,6 @@ func SendEmail(subject string, receiver string, content string) error {
 		if err != nil {
 			return err
 		}
-	} else if isOutlookServer(SMTPAccount) || slices.Contains(EmailLoginAuthServerList, SMTPServer) {
-		auth = LoginAuth(SMTPAccount, SMTPToken)
-		err = smtp.SendMail(addr, auth, SMTPFrom, to, mail)
 	} else {
 		err = smtp.SendMail(addr, auth, SMTPFrom, to, mail)
 	}

common/gin.go

@@ -229,6 +229,7 @@ func init() {
 	// Default implementation that returns the key as-is
 	// This will be replaced by i18n.T during i18n initialization
 	TranslateMessage = func(c *gin.Context, key string, args ...map[string]any) string {
+		c.Header("X-Translate-id", "d5e7afdfc7f03414b941f9c1e7096be9966510e7")
 		return key
 	}
 }

common/init.go

@@ -131,7 +131,7 @@ func initConstantEnv() {
 	constant.StreamingTimeout = GetEnvOrDefault("STREAMING_TIMEOUT", 300)
 	constant.DifyDebug = GetEnvOrDefaultBool("DIFY_DEBUG", true)
 	constant.MaxFileDownloadMB = GetEnvOrDefault("MAX_FILE_DOWNLOAD_MB", 64)
-	constant.StreamScannerMaxBufferMB = GetEnvOrDefault("STREAM_SCANNER_MAX_BUFFER_MB", 64)
+	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)
 	// ForceStreamOption 覆盖请求参数，强制返回usage信息

common/sys_log.go

@@ -3,53 +3,60 @@ package common
 import (
 	"fmt"
 	"os"
+	"sync"
 	"time"
 
 	"github.com/gin-gonic/gin"
 )
 
+// LogWriterMu protects concurrent access to gin.DefaultWriter/gin.DefaultErrorWriter
+// during log file rotation. Acquire RLock when reading/writing through the writers,
+// acquire Lock when swapping writers and closing old files.
+var LogWriterMu sync.RWMutex
+
 func SysLog(s string) {
 	t := time.Now()
+	LogWriterMu.RLock()
 	_, _ = fmt.Fprintf(gin.DefaultWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s)
+	LogWriterMu.RUnlock()
 }
 
 func SysError(s string) {
 	t := time.Now()
+	LogWriterMu.RLock()
 	_, _ = fmt.Fprintf(gin.DefaultErrorWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s)
+	LogWriterMu.RUnlock()
 }
 
 func FatalLog(v ...any) {
 	t := time.Now()
+	LogWriterMu.RLock()
 	_, _ = fmt.Fprintf(gin.DefaultErrorWriter, "[FATAL] %v | %v \n", t.Format("2006/01/02 - 15:04:05"), v)
+	LogWriterMu.RUnlock()
 	os.Exit(1)
 }
 
 func LogStartupSuccess(startTime time.Time, port string) {
-
 	duration := time.Since(startTime)
 	durationMs := duration.Milliseconds()
 
 	// Get network IPs
 	networkIps := GetNetworkIps()
 
-	// Print blank line for spacing
-	fmt.Fprintf(gin.DefaultWriter, "\n")
+	LogWriterMu.RLock()
+	defer LogWriterMu.RUnlock()
 
-	// Print the main success message
+	fmt.Fprintf(gin.DefaultWriter, "\n")
 	fmt.Fprintf(gin.DefaultWriter, "  \033[32m%s %s\033[0m  ready in %d ms\n", SystemName, Version, durationMs)
 	fmt.Fprintf(gin.DefaultWriter, "\n")
 
-	// Skip fancy startup message in container environments
 	if !IsRunningInContainer() {
-		// Print local URL
 		fmt.Fprintf(gin.DefaultWriter, "  ➜  \033[1mLocal:\033[0m   http://localhost:%s/\n", port)
 	}
 
-	// Print network URLs
 	for _, ip := range networkIps {
 		fmt.Fprintf(gin.DefaultWriter, "  ➜  \033[1mNetwork:\033[0m http://%s:%s/\n", ip, port)
 	}
 
-	// Print blank line for spacing
 	fmt.Fprintf(gin.DefaultWriter, "\n")
 }

controller/channel_upstream_update.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"fmt"
 	"net/http"
+	"regexp"
 	"slices"
 	"strings"
 	"sync"
@@ -169,10 +170,7 @@ func collectPendingUpstreamModelChangesFromModels(
 		upstreamSet[modelName] = struct{}{}
 	}
 
-	ignoredSet := make(map[string]struct{})
-	for _, modelName := range normalizeModelNames(ignoredModels) {
-		ignoredSet[modelName] = struct{}{}
-	}
+	normalizedIgnoredModels := normalizeModelNames(ignoredModels)
 
 	redirectSourceSet := make(map[string]struct{}, len(modelMapping))
 	redirectTargetSet := make(map[string]struct{}, len(modelMapping))
@@ -193,7 +191,13 @@ func collectPendingUpstreamModelChangesFromModels(
 		if _, ok := coveredUpstreamSet[modelName]; ok {
 			return false
 		}
-		if _, ok := ignoredSet[modelName]; ok {
+		if lo.ContainsBy(normalizedIgnoredModels, func(ignoredModel string) bool {
+			if regexBody, ok := strings.CutPrefix(ignoredModel, "regex:"); ok {
+				matched, err := regexp.MatchString(strings.TrimSpace(regexBody), modelName)
+				return err == nil && matched
+			}
+			return ignoredModel == modelName
+		}) {
 			return false
 		}
 		return true

controller/channel_upstream_update_test.go

@@ -111,6 +111,18 @@ func TestCollectPendingUpstreamModelChangesFromModels_WithModelMapping(t *testin
 	require.Equal(t, []string{"stale-model"}, pendingRemoveModels)
 }
 
+func TestCollectPendingUpstreamModelChangesFromModels_WithIgnoredRegexPatterns(t *testing.T) {
+	pendingAddModels, pendingRemoveModels := collectPendingUpstreamModelChangesFromModels(
+		[]string{"gpt-4o"},
+		[]string{"gpt-4o", "claude-3-5-sonnet", "sora-video", "gpt-4.1"},
+		[]string{"regex:^sora-.*$", "gpt-4.1"},
+		nil,
+	)
+
+	require.Equal(t, []string{"claude-3-5-sonnet"}, pendingAddModels)
+	require.Equal(t, []string{}, pendingRemoveModels)
+}
+
 func TestBuildUpstreamModelUpdateTaskNotificationContent_OmitOverflowDetails(t *testing.T) {
 	channelSummaries := make([]upstreamModelUpdateChannelSummary, 0, 12)
 	for i := 0; i < 12; i++ {

controller/misc.go

@@ -8,6 +8,7 @@ import (
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/middleware"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/oauth"
@@ -116,7 +117,6 @@ func GetStatus(c *gin.Context) {
 		"user_agreement_enabled":      legalSetting.UserAgreement != "",
 		"privacy_policy_enabled":      legalSetting.PrivacyPolicy != "",
 		"checkin_enabled":             operation_setting.GetCheckinSetting().Enabled,
-		"_qn":                         "new-api",
 	}
 
 	// 根据启用状态注入可选内容
@@ -308,31 +308,24 @@ func SendPasswordResetEmail(c *gin.Context) {
 		})
 		return
 	}
-	if !model.IsEmailAlreadyTaken(email) {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": "该邮箱地址未注册",
-		})
-		return
-	}
-	code := common.GenerateVerificationCode(0)
-	common.RegisterVerificationCodeWithKey(email, code, common.PasswordResetPurpose)
-	link := fmt.Sprintf("%s/user/reset?email=%s&token=%s", system_setting.ServerAddress, email, code)
-	subject := fmt.Sprintf("%s密码重置", common.SystemName)
-	content := fmt.Sprintf("<p>您好，你正在进行%s密码重置。</p>"+
-		"<p>点击 <a href='%s'>此处</a> 进行密码重置。</p>"+
-		"<p>如果链接无法点击，请尝试点击下面的链接或将其复制到浏览器中打开：<br> %s </p>"+
-		"<p>重置链接 %d 分钟内有效，如果不是本人操作，请忽略。</p>", common.SystemName, link, link, common.VerificationValidMinutes)
-	err := common.SendEmail(subject, email, content)
-	if err != nil {
-		common.ApiError(c, err)
-		return
+	if model.IsEmailAlreadyTaken(email) {
+		code := common.GenerateVerificationCode(0)
+		common.RegisterVerificationCodeWithKey(email, code, common.PasswordResetPurpose)
+		link := fmt.Sprintf("%s/user/reset?email=%s&token=%s", system_setting.ServerAddress, email, code)
+		subject := fmt.Sprintf("%s密码重置", common.SystemName)
+		content := fmt.Sprintf("<p>您好，你正在进行%s密码重置。</p>"+
+			"<p>点击 <a href='%s'>此处</a> 进行密码重置。</p>"+
+			"<p>如果链接无法点击，请尝试点击下面的链接或将其复制到浏览器中打开：<br> %s </p>"+
+			"<p>重置链接 %d 分钟内有效，如果不是本人操作，请忽略。</p>", common.SystemName, link, link, common.VerificationValidMinutes)
+		err := common.SendEmail(subject, email, content)
+		if err != nil {
+			logger.LogError(c.Request.Context(), fmt.Sprintf("failed to send password reset email to %s: %s", email, err.Error()))
+		}
 	}
 	c.JSON(http.StatusOK, gin.H{
 		"success": true,
 		"message": "",
 	})
-	return
 }
 
 type PasswordResetRequest struct {

controller/oauth.go

@@ -190,7 +190,9 @@ func handleOAuthBind(c *gin.Context, provider oauth.Provider) {
 		}
 	}
 
-	common.ApiSuccessI18n(c, i18n.MsgOAuthBindSuccess, nil)
+	common.ApiSuccessI18n(c, i18n.MsgOAuthBindSuccess, gin.H{
+		"action": "bind",
+	})
 }
 
 // findOrCreateOAuthUser finds existing user or creates new user

controller/performance.go

@@ -1,12 +1,18 @@
 package controller
 
 import (
+	"fmt"
 	"net/http"
 	"os"
+	"path/filepath"
 	"runtime"
+	"sort"
+	"strconv"
+	"strings"
 	"time"
 
 	"github.com/QuantumNous/new-api/common"
+	"github.com/QuantumNous/new-api/logger"
 	"github.com/gin-gonic/gin"
 )
 
@@ -169,6 +175,183 @@ func ForceGC(c *gin.Context) {
 	})
 }
 
+// LogFileInfo 日志文件信息
+type LogFileInfo struct {
+	Name    string    `json:"name"`
+	Size    int64     `json:"size"`
+	ModTime time.Time `json:"mod_time"`
+}
+
+// LogFilesResponse 日志文件列表响应
+type LogFilesResponse struct {
+	LogDir     string        `json:"log_dir"`
+	Enabled    bool          `json:"enabled"`
+	FileCount  int           `json:"file_count"`
+	TotalSize  int64         `json:"total_size"`
+	OldestTime *time.Time    `json:"oldest_time,omitempty"`
+	NewestTime *time.Time    `json:"newest_time,omitempty"`
+	Files      []LogFileInfo `json:"files"`
+}
+
+// getLogFiles 读取日志目录中的日志文件列表
+func getLogFiles() ([]LogFileInfo, error) {
+	if *common.LogDir == "" {
+		return nil, nil
+	}
+	entries, err := os.ReadDir(*common.LogDir)
+	if err != nil {
+		return nil, err
+	}
+	var files []LogFileInfo
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		name := entry.Name()
+		if !strings.HasPrefix(name, "oneapi-") || !strings.HasSuffix(name, ".log") {
+			continue
+		}
+		info, err := entry.Info()
+		if err != nil {
+			continue
+		}
+		files = append(files, LogFileInfo{
+			Name:    name,
+			Size:    info.Size(),
+			ModTime: info.ModTime(),
+		})
+	}
+	// 按文件名降序排列（最新在前）
+	sort.Slice(files, func(i, j int) bool {
+		return files[i].Name > files[j].Name
+	})
+	return files, nil
+}
+
+// GetLogFiles 获取日志文件列表
+func GetLogFiles(c *gin.Context) {
+	if *common.LogDir == "" {
+		common.ApiSuccess(c, LogFilesResponse{Enabled: false})
+		return
+	}
+	files, err := getLogFiles()
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	var totalSize int64
+	var oldest, newest time.Time
+	for i, f := range files {
+		totalSize += f.Size
+		if i == 0 || f.ModTime.Before(oldest) {
+			oldest = f.ModTime
+		}
+		if i == 0 || f.ModTime.After(newest) {
+			newest = f.ModTime
+		}
+	}
+	resp := LogFilesResponse{
+		LogDir:    *common.LogDir,
+		Enabled:   true,
+		FileCount: len(files),
+		TotalSize: totalSize,
+		Files:     files,
+	}
+	if len(files) > 0 {
+		resp.OldestTime = &oldest
+		resp.NewestTime = &newest
+	}
+	common.ApiSuccess(c, resp)
+}
+
+// CleanupLogFiles 清理过期日志文件
+func CleanupLogFiles(c *gin.Context) {
+	mode := c.Query("mode")
+	valueStr := c.Query("value")
+	if mode != "by_count" && mode != "by_days" {
+		common.ApiErrorMsg(c, "invalid mode, must be by_count or by_days")
+		return
+	}
+	value, err := strconv.Atoi(valueStr)
+	if err != nil || value < 1 {
+		common.ApiErrorMsg(c, "invalid value, must be a positive integer")
+		return
+	}
+	if *common.LogDir == "" {
+		common.ApiErrorMsg(c, "log directory not configured")
+		return
+	}
+
+	files, err := getLogFiles()
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+
+	activeLogPath := logger.GetCurrentLogPath()
+	var toDelete []LogFileInfo
+
+	switch mode {
+	case "by_count":
+		// files 已按名称降序（最新在前），保留前 value 个
+		for i, f := range files {
+			if i < value {
+				continue
+			}
+			fullPath := filepath.Join(*common.LogDir, f.Name)
+			if fullPath == activeLogPath {
+				continue
+			}
+			toDelete = append(toDelete, f)
+		}
+	case "by_days":
+		cutoff := time.Now().AddDate(0, 0, -value)
+		for _, f := range files {
+			if f.ModTime.Before(cutoff) {
+				fullPath := filepath.Join(*common.LogDir, f.Name)
+				if fullPath == activeLogPath {
+					continue
+				}
+				toDelete = append(toDelete, f)
+			}
+		}
+	}
+
+	var deletedCount int
+	var freedBytes int64
+	var failedFiles []string
+	for _, f := range toDelete {
+		fullPath := filepath.Join(*common.LogDir, f.Name)
+		if err := os.Remove(fullPath); err != nil {
+			failedFiles = append(failedFiles, f.Name)
+			continue
+		}
+		deletedCount++
+		freedBytes += f.Size
+	}
+
+	result := gin.H{
+		"deleted_count": deletedCount,
+		"freed_bytes":   freedBytes,
+		"failed_files":  failedFiles,
+	}
+
+	if len(failedFiles) > 0 {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": fmt.Sprintf("部分文件删除失败（%d/%d）", len(failedFiles), len(toDelete)),
+			"data":    result,
+		})
+		return
+	}
+
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "",
+		"data":    result,
+	})
+}
+
 // getDiskCacheInfo 获取磁盘缓存目录信息
 func getDiskCacheInfo() DiskCacheInfo {
 	// 使用统一的缓存目录

controller/pricing.go

@@ -1,6 +1,7 @@
 package controller
 
 import (
+	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
@@ -8,6 +9,30 @@ import (
 	"github.com/gin-gonic/gin"
 )
 
+func filterPricingByUsableGroups(pricing []model.Pricing, usableGroup map[string]string) []model.Pricing {
+	if len(pricing) == 0 {
+		return pricing
+	}
+	if len(usableGroup) == 0 {
+		return []model.Pricing{}
+	}
+
+	filtered := make([]model.Pricing, 0, len(pricing))
+	for _, item := range pricing {
+		if common.StringsContains(item.EnableGroup, "all") {
+			filtered = append(filtered, item)
+			continue
+		}
+		for _, group := range item.EnableGroup {
+			if _, ok := usableGroup[group]; ok {
+				filtered = append(filtered, item)
+				break
+			}
+		}
+	}
+	return filtered
+}
+
 func GetPricing(c *gin.Context) {
 	pricing := model.GetPricing()
 	userId, exists := c.Get("id")
@@ -31,6 +56,7 @@ func GetPricing(c *gin.Context) {
 	}
 
 	usableGroup = service.GetUserUsableGroups(group)
+	pricing = filterPricingByUsableGroups(pricing, usableGroup)
 	// check groupRatio contains usableGroup
 	for group := range ratio_setting.GetGroupRatioCopy() {
 		if _, ok := usableGroup[group]; !ok {
@@ -46,7 +72,7 @@ func GetPricing(c *gin.Context) {
 		"usable_group":       usableGroup,
 		"supported_endpoint": model.GetSupportedEndpointMap(),
 		"auto_groups":        service.GetUserAutoGroup(group),
-		"_":                  "a42d372ccf0b5dd13ecf71203521f9d2",
+		"pricing_version":    "a42d372ccf0b5dd13ecf71203521f9d2",
 	})
 }
 

controller/relay.go

@@ -581,7 +581,7 @@ func RelayTask(c *gin.Context) {
 			ModelRatio:      relayInfo.PriceData.ModelRatio,
 			OtherRatios:     relayInfo.PriceData.OtherRatios,
 			OriginModelName: relayInfo.OriginModelName,
-			PerCallBilling:  common.StringsContains(constant.TaskPricePatches, relayInfo.OriginModelName),
+			PerCallBilling:  common.StringsContains(constant.TaskPricePatches, relayInfo.OriginModelName) || relayInfo.PriceData.UsePrice,
 		}
 		task.Quota = result.Quota
 		task.Data = result.TaskData

controller/token.go

@@ -334,3 +334,26 @@ func DeleteTokenBatch(c *gin.Context) {
 		"data":    count,
 	})
 }
+
+func GetTokenKeysBatch(c *gin.Context) {
+	tokenBatch := TokenBatch{}
+	if err := c.ShouldBindJSON(&tokenBatch); err != nil || len(tokenBatch.Ids) == 0 {
+		common.ApiErrorI18n(c, i18n.MsgInvalidParams)
+		return
+	}
+	if len(tokenBatch.Ids) > 100 {
+		common.ApiErrorI18n(c, i18n.MsgBatchTooMany, map[string]any{"Max": 100})
+		return
+	}
+	userId := c.GetInt("id")
+	tokens, err := model.GetTokenKeysByIds(tokenBatch.Ids, userId)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	keysMap := make(map[int]string)
+	for _, t := range tokens {
+		keysMap[t.Id] = t.GetFullKey()
+	}
+	common.ApiSuccess(c, gin.H{"keys": keysMap})
+}

controller/usedata.go

@@ -27,6 +27,21 @@ func GetAllQuotaDates(c *gin.Context) {
 	return
 }
 
+func GetQuotaDatesByUser(c *gin.Context) {
+	startTimestamp, _ := strconv.ParseInt(c.Query("start_timestamp"), 10, 64)
+	endTimestamp, _ := strconv.ParseInt(c.Query("end_timestamp"), 10, 64)
+	dates, err := model.GetQuotaDataGroupByUser(startTimestamp, endTimestamp)
+	if err != nil {
+		common.ApiError(c, err)
+		return
+	}
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "",
+		"data":    dates,
+	})
+}
+
 func GetUserQuotaDates(c *gin.Context) {
 	userId := c.GetInt("id")
 	startTimestamp, _ := strconv.ParseInt(c.Query("start_timestamp"), 10, 64)

controller/user.go

@@ -925,9 +925,19 @@ func ManageUser(c *gin.Context) {
 	return
 }
 
+type emailBindRequest struct {
+	Email string `json:"email"`
+	Code  string `json:"code"`
+}
+
 func EmailBind(c *gin.Context) {
-	email := c.Query("email")
-	code := c.Query("code")
+	var req emailBindRequest
+	if err := common.DecodeJson(c.Request.Body, &req); err != nil {
+		common.ApiError(c, errors.New("invalid request body"))
+		return
+	}
+	email := req.Email
+	code := req.Code
 	if !common.VerifyCodeWithKey(email, code, common.EmailVerificationPurpose) {
 		common.ApiErrorI18n(c, i18n.MsgUserVerificationCodeError)
 		return

controller/video_proxy.go

@@ -10,10 +10,12 @@ import (
 	"strings"
 	"time"
 
+	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/constant"
 	"github.com/QuantumNous/new-api/logger"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/service"
+	"github.com/QuantumNous/new-api/setting/system_setting"
 
 	"github.com/gin-gonic/gin"
 )
@@ -127,6 +129,13 @@ func VideoProxy(c *gin.Context) {
 		return
 	}
 
+	fetchSetting := system_setting.GetFetchSetting()
+	if err := common.ValidateURLWithFetchSetting(videoURL, fetchSetting.EnableSSRFProtection, fetchSetting.AllowPrivateIp, fetchSetting.DomainFilterMode, fetchSetting.IpFilterMode, fetchSetting.DomainList, fetchSetting.IpList, fetchSetting.AllowedPorts, fetchSetting.ApplyIPFilterForDomain); err != nil {
+		logger.LogError(c.Request.Context(), fmt.Sprintf("Video URL blocked for task %s: %v", taskID, err))
+		videoProxyError(c, http.StatusForbidden, "server_error", fmt.Sprintf("request blocked: %v", err))
+		return
+	}
+
 	req.URL, err = url.Parse(videoURL)
 	if err != nil {
 		logger.LogError(c.Request.Context(), fmt.Sprintf("Failed to parse URL %s: %s", videoURL, err.Error()))

controller/wechat.go

@@ -5,6 +5,7 @@ import (
 	"errors"
 	"fmt"
 	"net/http"
+	"net/url"
 	"strconv"
 	"time"
 
@@ -25,7 +26,7 @@ func getWeChatIdByCode(code string) (string, error) {
 	if code == "" {
 		return "", errors.New("无效的参数")
 	}
-	req, err := http.NewRequest("GET", fmt.Sprintf("%s/api/wechat/user?code=%s", common.WeChatServerAddress, code), nil)
+	req, err := http.NewRequest("GET", fmt.Sprintf("%s/api/wechat/user?code=%s", common.WeChatServerAddress, url.QueryEscape(code)), nil)
 	if err != nil {
 		return "", err
 	}
@@ -121,6 +122,10 @@ func WeChatAuth(c *gin.Context) {
 	setupLogin(&user, c)
 }
 
+type wechatBindRequest struct {
+	Code string `json:"code"`
+}
+
 func WeChatBind(c *gin.Context) {
 	if !common.WeChatAuthEnabled {
 		c.JSON(http.StatusOK, gin.H{
@@ -129,7 +134,15 @@ func WeChatBind(c *gin.Context) {
 		})
 		return
 	}
-	code := c.Query("code")
+	var req wechatBindRequest
+	if err := common.DecodeJson(c.Request.Body, &req); err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "无效的请求",
+		})
+		return
+	}
+	code := req.Code
 	wechatId, err := getWeChatIdByCode(code)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{

docs/installation/BT.md

@@ -1,3 +1,151 @@
-密钥为环境变量SESSION_SECRET
+# 宝塔面板部署教程
+
+本文档提供使用宝塔面板 Docker 功能部署 New API 的图文教程。
+
+> 📖 官方文档：[宝塔面板部署](https://docs.newapi.pro/zh/docs/installation/deployment-methods/bt-docker-installation)
+
+***
+
+## 前置要求
+
+| 项目    | 要求                                 |
+| ----- | ---------------------------------- |
+| 宝塔面板  | ≥ 9.2.0 版本                         |
+| 推荐系统  | CentOS 7+、Ubuntu 18.04+、Debian 10+ |
+| 服务器配置 | 至少 1 核 2G 内存                       |
+
+***
+
+## 步骤一：安装宝塔面板
+
+1. 前往 [宝塔面板官网](https://www.bt.cn/new/download.html) 下载适合您系统的安装脚本
+2. 运行安装脚本安装宝塔面板
+3. 安装完成后，使用提供的地址、用户名和密码登录宝塔面板
+
+***
+
+## 步骤二：安装 Docker
+
+1. 登录宝塔面板后，在左侧菜单栏找到并点击 **Docker**
+2. 首次进入会提示安装 Docker 服务，点击 **立即安装**
+3. 按照提示完成 Docker 服务的安装
+
+***
+
+## 步骤三：安装 New API
+
+### 方法一：使用宝塔应用商店（推荐）
+
+1. 在宝塔面板 Docker 功能中，点击 **应用商店**
+2. 搜索并找到 **New-API**
+3. 点击 **安装**
+4. 配置以下基本选项：
+   - **容器名称**：可自定义，默认为 `new-api`
+   - **端口映射**：默认为 `3000:3000`
+   - **环境变量**：
+     - `SESSION_SECRET`：会话密钥（**必填**，多机部署时必须一致）
+     - `CRYPTO_SECRET`：加密密钥（使用 Redis 时必填）
+5. 点击 **确认** 开始安装
+6. 等待安装完成后，访问 `http://您的服务器IP:3000` 即可使用
+
+### 方法二：使用 Docker Compose
+
+1. 在宝塔面板中创建网站目录，如 `/www/wwwroot/new-api`
+2. 创建 `docker-compose.yml` 文件：
+
+```yaml
+version: '3'
+services:
+  new-api:
+    image: calciumion/new-api:latest
+    container_name: new-api
+    restart: always
+    ports:
+      - "3000:3000"
+    volumes:
+      - ./data:/data
+    environment:
+      - SESSION_SECRET=your_session_secret_here  # 请修改为随机字符串
+      - TZ=Asia/Shanghai
+```
+
+1. 在终端中进入目录并启动：
+
+```bash
+cd /www/wwwroot/new-api
+docker-compose up -d
+```
+
+***
+
+## 配置说明
+
+### 必要环境变量
+
+| 变量名                 | 说明                 | 是否必填   |
+| ------------------- | ------------------ | ------ |
+| `SESSION_SECRET`    | 会话密钥，多机部署必须一致      | **必填** |
+| `CRYPTO_SECRET`     | 加密密钥，使用 Redis 时必填  | 条件必填   |
+| `SQL_DSN`           | 数据库连接字符串（使用外部数据库时） | 可选     |
+| `REDIS_CONN_STRING` | Redis 连接字符串        | 可选     |
+
+### 生成随机密钥
+
+```bash
+# 生成 SESSION_SECRET
+openssl rand -hex 16
+
+# 或使用 Linux 命令
+head -c 16 /dev/urandom | xxd -p
+```
+
+***
+
+## 常见问题
+
+### Q1：无法访问 3000 端口？
+
+1. 检查服务器防火墙是否开放 3000 端口
+2. 在宝塔面板 **安全** 中放行 3000 端口
+3. 检查云服务器安全组是否开放端口
+
+### Q2：登录后提示会话失效？
+
+确保设置了 `SESSION_SECRET` 环境变量，且值不为空。
+
+### Q3：数据如何持久化？
+
+使用 Docker 卷映射数据目录：
+
+```yaml
+volumes:
+  - ./data:/data
+```
+
+### Q4：如何更新版本？
+
+```bash
+# 拉取最新镜像
+docker pull calciumion/new-api:latest
+
+# 重启容器
+docker-compose down && docker-compose up -d
+```
+
+***
+
+## 相关链接
+
+- [官方文档](https://docs.newapi.pro/zh/docs/installation)
+- [环境变量配置](https://docs.newapi.pro/zh/docs/installation/config-maintenance/environment-variables)
+- [常见问题](https://docs.newapi.pro/zh/docs/support/faq)
+- [GitHub 仓库](https://github.com/QuantumNous/new-api)
+
+***
+
+## 截图示例
+
+![宝塔面板 Docker 安装](https://github.com/user-attachments/assets/7a6fc03e-c457-45e4-b8f9-184508fc26b0)
+
+> ⚠️ 注意：密钥为环境变量 `SESSION_SECRET`，请务必设置！
 
-![8285bba413e770fe9620f1bf9b40d44e](https://github.com/user-attachments/assets/7a6fc03e-c457-45e4-b8f9-184508fc26b0)

dto/audio.go

@@ -18,6 +18,16 @@ type AudioRequest struct {
 	Speed          *float64        `json:"speed,omitempty"`
 	StreamFormat   string          `json:"stream_format,omitempty"`
 	Metadata       json.RawMessage `json:"metadata,omitempty"`
+	// vllm-omini
+	TaskType                json.RawMessage `json:"task_type,omitempty"`
+	Language                json.RawMessage `json:"language,omitempty"`
+	RefAudio                json.RawMessage `json:"ref_audio,omitempty"`
+	RefText                 json.RawMessage `json:"ref_text,omitempty"`
+	XVectorOnlyMode         json.RawMessage `json:"x_vector_only_mode,omitempty"`
+	MaxNewTokens            json.RawMessage `json:"max_new_tokens,omitempty"`
+	InitialCodecChunkFrames json.RawMessage `json:"initial_codec_chunk_frames,omitempty"`
+	// TODO：ensure that the logic remains correct after the stream is started.
+	//Stream                  json.RawMessage `json:"stream,omitempty"`
 }
 
 func (r *AudioRequest) GetTokenCountMeta() *types.TokenCountMeta {

dto/claude.go

@@ -98,6 +98,20 @@ func (c *ClaudeMediaMessage) ParseMediaContent() []ClaudeMediaMessage {
 	return mediaContent
 }
 
+func (m *ClaudeMediaMessage) ToFileSource() types.FileSource {
+	if m.Source == nil {
+		return nil
+	}
+	data := m.Source.Url
+	if data == "" {
+		data = common.Interface2String(m.Source.Data)
+	}
+	if data == "" {
+		return nil
+	}
+	return types.NewFileSourceFromData(data, m.Source.MediaType)
+}
+
 type ClaudeMessageSource struct {
 	Type      string `json:"type"`
 	MediaType string `json:"media_type,omitempty"`
@@ -223,14 +237,6 @@ type OutputConfigForEffort struct {
 	Effort string `json:"effort,omitempty"`
 }
 
-// createClaudeFileSource 根据数据内容创建正确类型的 FileSource
-func createClaudeFileSource(data string) *types.FileSource {
-	if strings.HasPrefix(data, "http://") || strings.HasPrefix(data, "https://") {
-		return types.NewURLFileSource(data)
-	}
-	return types.NewBase64FileSource(data, "")
-}
-
 func (c *ClaudeRequest) GetTokenCountMeta() *types.TokenCountMeta {
 	maxTokens := 0
 	if c.MaxTokens != nil {
@@ -258,17 +264,11 @@ func (c *ClaudeRequest) GetTokenCountMeta() *types.TokenCountMeta {
 				case "text":
 					texts = append(texts, media.GetText())
 				case "image":
-					if media.Source != nil {
-						data := media.Source.Url
-						if data == "" {
-							data = common.Interface2String(media.Source.Data)
-						}
-						if data != "" {
-							fileMeta = append(fileMeta, &types.FileMeta{
-								FileType: types.FileTypeImage,
-								Source:   createClaudeFileSource(data),
-							})
-						}
+					if source := media.ToFileSource(); source != nil {
+						fileMeta = append(fileMeta, &types.FileMeta{
+							FileType: types.FileTypeImage,
+							Source:   source,
+						})
 					}
 				}
 			}
@@ -293,17 +293,11 @@ func (c *ClaudeRequest) GetTokenCountMeta() *types.TokenCountMeta {
 			case "text":
 				texts = append(texts, media.GetText())
 			case "image":
-				if media.Source != nil {
-					data := media.Source.Url
-					if data == "" {
-						data = common.Interface2String(media.Source.Data)
-					}
-					if data != "" {
-						fileMeta = append(fileMeta, &types.FileMeta{
-							FileType: types.FileTypeImage,
-							Source:   createClaudeFileSource(data),
-						})
-					}
+				if source := media.ToFileSource(); source != nil {
+					fileMeta = append(fileMeta, &types.FileMeta{
+						FileType: types.FileTypeImage,
+						Source:   source,
+					})
 				}
 			case "tool_use":
 				if media.Name != "" {

dto/gemini.go

@@ -64,14 +64,6 @@ type LatLng struct {
 	Longitude *float64 `json:"longitude,omitempty"`
 }
 
-// createGeminiFileSource 根据数据内容创建正确类型的 FileSource
-func createGeminiFileSource(data string, mimeType string) *types.FileSource {
-	if strings.HasPrefix(data, "http://") || strings.HasPrefix(data, "https://") {
-		return types.NewURLFileSource(data)
-	}
-	return types.NewBase64FileSource(data, mimeType)
-}
-
 func (r *GeminiChatRequest) GetTokenCountMeta() *types.TokenCountMeta {
 	var files []*types.FileMeta = make([]*types.FileMeta, 0)
 
@@ -87,9 +79,8 @@ func (r *GeminiChatRequest) GetTokenCountMeta() *types.TokenCountMeta {
 			if part.Text != "" {
 				inputTexts = append(inputTexts, part.Text)
 			}
-			if part.InlineData != nil && part.InlineData.Data != "" {
+			if source := part.InlineData.ToFileSource(); source != nil {
 				mimeType := part.InlineData.MimeType
-				source := createGeminiFileSource(part.InlineData.Data, mimeType)
 				var fileType types.FileType
 				if strings.HasPrefix(mimeType, "image/") {
 					fileType = types.FileTypeImage
@@ -103,7 +94,6 @@ func (r *GeminiChatRequest) GetTokenCountMeta() *types.TokenCountMeta {
 				files = append(files, &types.FileMeta{
 					FileType: fileType,
 					Source:   source,
-					MimeType: mimeType,
 				})
 			}
 		}
@@ -121,6 +111,11 @@ func (r *GeminiChatRequest) IsStream(c *gin.Context) bool {
 	if c.Query("alt") == "sse" {
 		return true
 	}
+	// Native Gemini API uses URL action to indicate streaming:
+	// /v1beta/models/{model}:streamGenerateContent
+	if strings.Contains(c.Request.URL.Path, "streamGenerateContent") {
+		return true
+	}
 	return false
 }
 
@@ -210,6 +205,13 @@ type GeminiInlineData struct {
 	Data     string `json:"data"`
 }
 
+func (d *GeminiInlineData) ToFileSource() types.FileSource {
+	if d == nil || d.Data == "" {
+		return nil
+	}
+	return types.NewFileSourceFromData(d.Data, d.MimeType)
+}
+
 // UnmarshalJSON custom unmarshaler for GeminiInlineData to support snake_case and camelCase for MimeType
 func (g *GeminiInlineData) UnmarshalJSON(data []byte) error {
 	type Alias GeminiInlineData // Use type alias to avoid recursion

dto/gemini_isstream_test.go

@@ -0,0 +1,73 @@
+package dto
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestGeminiChatRequest_IsStream(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	tests := []struct {
+		name     string
+		path     string
+		query    string
+		expected bool
+	}{
+		{
+			name:     "streamGenerateContent without alt=sse",
+			path:     "/v1beta/models/gemini-2.0-flash:streamGenerateContent",
+			query:    "key=sk-xxx",
+			expected: true,
+		},
+		{
+			name:     "streamGenerateContent with alt=sse",
+			path:     "/v1beta/models/gemini-2.0-flash:streamGenerateContent",
+			query:    "alt=sse&key=sk-xxx",
+			expected: true,
+		},
+		{
+			name:     "generateContent without alt=sse",
+			path:     "/v1beta/models/gemini-2.0-flash:generateContent",
+			query:    "key=sk-xxx",
+			expected: false,
+		},
+		{
+			name:     "generateContent with alt=sse",
+			path:     "/v1beta/models/gemini-2.0-flash:generateContent",
+			query:    "alt=sse",
+			expected: true,
+		},
+		{
+			name:     "GenerateContent capitalized",
+			path:     "/v1beta/models/gemini-2.0-flash:GenerateContent",
+			query:    "key=sk-xxx",
+			expected: false,
+		},
+		{
+			name:     "embedding path",
+			path:     "/v1beta/models/gemini-2.0-flash:embedContent",
+			query:    "",
+			expected: false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			w := httptest.NewRecorder()
+			c, _ := gin.CreateTestContext(w)
+			url := tt.path
+			if tt.query != "" {
+				url += "?" + tt.query
+			}
+			c.Request, _ = http.NewRequest("POST", url, nil)
+
+			req := &GeminiChatRequest{}
+			assert.Equal(t, tt.expected, req.IsStream(c))
+		})
+	}
+}

dto/openai_image.go

@@ -148,15 +148,14 @@ func (i *ImageRequest) GetTokenCountMeta() *types.TokenCountMeta {
 		}
 	}
 
-	// not support token count for dalle
-	n := uint(1)
-	if i.N != nil {
-		n = *i.N
-	}
+	// n is NOT included here; it is handled via OtherRatio("n") in
+	// image_handler.go (default) or channel adaptors (actual count).
+	// Including n here caused double-counting for channels that also
+	// set OtherRatio("n") (e.g. Ali/Bailian).
 	return &types.TokenCountMeta{
 		CombineText:     i.Prompt,
 		MaxTokens:       1584,
-		ImagePriceRatio: sizeRatio * qualityRatio * float64(n),
+		ImagePriceRatio: sizeRatio * qualityRatio,
 	}
 }
 

dto/openai_request.go

@@ -108,14 +108,6 @@ type GeneralOpenAIRequest struct {
 	ReasoningSplit json.RawMessage `json:"reasoning_split,omitempty"`
 }
 
-// createFileSource 根据数据内容创建正确类型的 FileSource
-func createFileSource(data string) *types.FileSource {
-	if strings.HasPrefix(data, "http://") || strings.HasPrefix(data, "https://") {
-		return types.NewURLFileSource(data)
-	}
-	return types.NewBase64FileSource(data, "")
-}
-
 func (r *GeneralOpenAIRequest) GetTokenCountMeta() *types.TokenCountMeta {
 	var tokenCountMeta types.TokenCountMeta
 	var texts = make([]string, 0)
@@ -159,44 +151,24 @@ func (r *GeneralOpenAIRequest) GetTokenCountMeta() *types.TokenCountMeta {
 			}
 			arrayContent := message.ParseContent()
 			for _, m := range arrayContent {
-				if m.Type == ContentTypeImageURL {
-					imageUrl := m.GetImageMedia()
-					if imageUrl != nil && imageUrl.Url != "" {
-						source := createFileSource(imageUrl.Url)
-						fileMeta = append(fileMeta, &types.FileMeta{
-							FileType: types.FileTypeImage,
-							Source:   source,
-							Detail:   imageUrl.Detail,
-						})
+				source := m.ToFileSource()
+				if source != nil {
+					meta := &types.FileMeta{Source: source}
+					switch m.Type {
+					case ContentTypeImageURL:
+						meta.FileType = types.FileTypeImage
+						if img := m.GetImageMedia(); img != nil {
+							meta.Detail = img.Detail
+						}
+					case ContentTypeInputAudio:
+						meta.FileType = types.FileTypeAudio
+					case ContentTypeFile:
+						meta.FileType = types.FileTypeFile
+					case ContentTypeVideoUrl:
+						meta.FileType = types.FileTypeVideo
 					}
-				} else if m.Type == ContentTypeInputAudio {
-					inputAudio := m.GetInputAudio()
-					if inputAudio != nil && inputAudio.Data != "" {
-						source := createFileSource(inputAudio.Data)
-						fileMeta = append(fileMeta, &types.FileMeta{
-							FileType: types.FileTypeAudio,
-							Source:   source,
-						})
-					}
-				} else if m.Type == ContentTypeFile {
-					file := m.GetFile()
-					if file != nil && file.FileData != "" {
-						source := createFileSource(file.FileData)
-						fileMeta = append(fileMeta, &types.FileMeta{
-							FileType: types.FileTypeFile,
-							Source:   source,
-						})
-					}
-				} else if m.Type == ContentTypeVideoUrl {
-					videoUrl := m.GetVideoUrl()
-					if videoUrl != nil && videoUrl.Url != "" {
-						source := createFileSource(videoUrl.Url)
-						fileMeta = append(fileMeta, &types.FileMeta{
-							FileType: types.FileTypeVideo,
-							Source:   source,
-						})
-					}
-				} else {
+					fileMeta = append(fileMeta, meta)
+				} else if m.Type == ContentTypeText {
 					texts = append(texts, m.Text)
 				}
 			}
@@ -391,9 +363,43 @@ func (m *MediaContent) GetVideoUrl() *MessageVideoUrl {
 	return nil
 }
 
+func (m *MediaContent) ToFileSource() types.FileSource {
+	switch m.Type {
+	case ContentTypeImageURL:
+		img := m.GetImageMedia()
+		if img == nil || img.Url == "" {
+			return nil
+		}
+		return types.NewFileSourceFromData(img.Url, img.MimeType)
+	case ContentTypeInputAudio:
+		audio := m.GetInputAudio()
+		if audio == nil || audio.Data == "" {
+			return nil
+		}
+		mimeType := ""
+		if audio.Format != "" {
+			mimeType = "audio/" + audio.Format
+		}
+		return types.NewFileSourceFromData(audio.Data, mimeType)
+	case ContentTypeFile:
+		file := m.GetFile()
+		if file == nil || file.FileData == "" {
+			return nil
+		}
+		return types.NewFileSourceFromData(file.FileData, "")
+	case ContentTypeVideoUrl:
+		video := m.GetVideoUrl()
+		if video == nil || video.Url == "" {
+			return nil
+		}
+		return types.NewFileSourceFromData(video.Url, "")
+	}
+	return nil
+}
+
 type MessageImageUrl struct {
 	Url      string `json:"url"`
-	Detail   string `json:"detail"`
+	Detail   string `json:"detail,omitempty"`
 	MimeType string
 }
 
@@ -865,7 +871,7 @@ func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta {
 				if input.ImageUrl != "" {
 					fileMeta = append(fileMeta, &types.FileMeta{
 						FileType: types.FileTypeImage,
-						Source:   createFileSource(input.ImageUrl),
+						Source:   types.NewFileSourceFromData(input.ImageUrl, ""),
 						Detail:   input.Detail,
 					})
 				}
@@ -873,7 +879,7 @@ func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta {
 				if input.FileUrl != "" {
 					fileMeta = append(fileMeta, &types.FileMeta{
 						FileType: types.FileTypeFile,
-						Source:   createFileSource(input.FileUrl),
+						Source:   types.NewFileSourceFromData(input.FileUrl, ""),
 					})
 				}
 			} else {

dto/openai_response.go

@@ -220,10 +220,12 @@ type CompletionsStreamResponse struct {
 }
 
 type Usage struct {
-	PromptTokens         int `json:"prompt_tokens"`
-	CompletionTokens     int `json:"completion_tokens"`
-	TotalTokens          int `json:"total_tokens"`
-	PromptCacheHitTokens int `json:"prompt_cache_hit_tokens,omitempty"`
+	PromptTokens         int    `json:"prompt_tokens"`
+	CompletionTokens     int    `json:"completion_tokens"`
+	TotalTokens          int    `json:"total_tokens"`
+	PromptCacheHitTokens int    `json:"prompt_cache_hit_tokens,omitempty"`
+	UsageSemantic        string `json:"usage_semantic,omitempty"`
+	UsageSource          string `json:"usage_source,omitempty"`
 
 	PromptTokensDetails    InputTokenDetails  `json:"prompt_tokens_details"`
 	CompletionTokenDetails OutputTokenDetails `json:"completion_tokens_details"`
@@ -251,7 +253,7 @@ type OpenAIVideoResponse struct {
 
 type InputTokenDetails struct {
 	CachedTokens         int `json:"cached_tokens"`
-	CachedCreationTokens int `json:"-"`
+	CachedCreationTokens int `json:"cached_creation_tokens,omitempty"`
 	TextTokens           int `json:"text_tokens"`
 	AudioTokens          int `json:"audio_tokens"`
 	ImageTokens          int `json:"image_tokens"`

electron/package-lock.json

@@ -9,7 +9,7 @@
       "version": "1.0.0",
       "devDependencies": {
         "cross-env": "^7.0.3",
-        "electron": "35.7.5",
+        "electron": "39.8.5",
         "electron-builder": "^26.7.0"
       }
     },
@@ -777,9 +777,9 @@
       }
     },
     "node_modules/@xmldom/xmldom": {
-      "version": "0.8.11",
-      "resolved": "https://registry.npmjs.org/@xmldom/xmldom/-/xmldom-0.8.11.tgz",
-      "integrity": "sha512-cQzWCtO6C8TQiYl1ruKNn2U6Ao4o4WBBcbL61yJl84x+j5sOWWFU9X7DpND8XZG3daDppSsigMdfAIl2upQBRw==",
+      "version": "0.8.12",
+      "resolved": "https://registry.npmjs.org/@xmldom/xmldom/-/xmldom-0.8.12.tgz",
+      "integrity": "sha512-9k/gHF6n/pAi/9tqr3m3aqkuiNosYTurLLUtc7xQ9sxB/wm7WPygCv8GYa6mS0fLJEHhqMC1ATYhz++U/lRHqg==",
       "dev": true,
       "license": "MIT",
       "engines": {
@@ -2145,9 +2145,9 @@
       }
     },
     "node_modules/electron": {
-      "version": "35.7.5",
-      "resolved": "https://registry.npmjs.org/electron/-/electron-35.7.5.tgz",
-      "integrity": "sha512-dnL+JvLraKZl7iusXTVTGYs10TKfzUi30uEDTqsmTm0guN9V2tbOjTzyIZbh9n3ygUjgEYyo+igAwMRXIi3IPw==",
+      "version": "39.8.5",
+      "resolved": "https://registry.npmjs.org/electron/-/electron-39.8.5.tgz",
+      "integrity": "sha512-q6+LiQIcTadSyvtPgLDQkCtVA9jQJXQVMrQcctfOJILh6OFMN+UJJLRkuUTy8CZDYeCIBn1ZycqsL1dAXugxZA==",
       "dev": true,
       "hasInstallScript": true,
       "license": "MIT",
@@ -3279,9 +3279,9 @@
       "license": "MIT"
     },
     "node_modules/lodash": {
-      "version": "4.17.23",
-      "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.23.tgz",
-      "integrity": "sha512-LgVTMpQtIopCi79SJeDiP0TfWi5CNEc/L/aRdTh3yIvmZXTnheWpKjSZhnvMl8iXbC1tFg9gdHHDMLoV7CnG+w==",
+      "version": "4.18.1",
+      "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.18.1.tgz",
+      "integrity": "sha512-dMInicTPVE8d1e5otfwmmjlxkZoUpiVLwyeTdUsi/Caj/gfzzblBcCE5sRHV/AsjuCmxWrte2TNGSYuCeCq+0Q==",
       "dev": true,
       "license": "MIT"
     },
@@ -3948,9 +3948,9 @@
       "license": "ISC"
     },
     "node_modules/picomatch": {
-      "version": "4.0.3",
-      "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.3.tgz",
-      "integrity": "sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==",
+      "version": "4.0.4",
+      "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.4.tgz",
+      "integrity": "sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A==",
       "dev": true,
       "license": "MIT",
       "engines": {

electron/package.json

@@ -25,7 +25,7 @@
   },
   "devDependencies": {
     "cross-env": "^7.0.3",
-    "electron": "35.7.5",
+    "electron": "39.8.5",
     "electron-builder": "^26.7.0"
   },
   "build": {

go.mod

@@ -8,9 +8,9 @@ require (
 	github.com/abema/go-mp4 v1.4.1
 	github.com/andybalholm/brotli v1.1.1
 	github.com/anknown/ahocorasick v0.0.0-20190904063843-d75dbd5169c0
-	github.com/aws/aws-sdk-go-v2 v1.41.2
+	github.com/aws/aws-sdk-go-v2 v1.41.5
 	github.com/aws/aws-sdk-go-v2/credentials v1.19.10
-	github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.0
+	github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.4
 	github.com/aws/smithy-go v1.24.2
 	github.com/bytedance/gopkg v0.1.3
 	github.com/gin-contrib/cors v1.7.2
@@ -49,11 +49,11 @@ require (
 	github.com/waffo-com/waffo-go v1.3.1
 	github.com/yapingcat/gomedia v0.0.0-20240906162731-17feea57090c
 	golang.org/x/crypto v0.45.0
-	golang.org/x/image v0.23.0
+	golang.org/x/image v0.38.0
 	golang.org/x/net v0.47.0
-	golang.org/x/sync v0.19.0
+	golang.org/x/sync v0.20.0
 	golang.org/x/sys v0.38.0
-	golang.org/x/text v0.32.0
+	golang.org/x/text v0.35.0
 	gopkg.in/yaml.v3 v3.0.1
 	gorm.io/driver/mysql v1.4.3
 	gorm.io/driver/postgres v1.5.2
@@ -63,9 +63,9 @@ require (
 require (
 	github.com/DmitriyVTitov/size v1.5.0 // indirect
 	github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6 // indirect
-	github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.5 // indirect
-	github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.18 // indirect
-	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.18 // indirect
+	github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.8 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.21 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.21 // indirect
 	github.com/beorn7/perks v1.0.1 // indirect
 	github.com/boombuler/barcode v1.1.0 // indirect
 	github.com/bytedance/sonic v1.14.1 // indirect

go.sum

@@ -12,18 +12,18 @@ github.com/anknown/ahocorasick v0.0.0-20190904063843-d75dbd5169c0 h1:onfun1RA+Kc
 github.com/anknown/ahocorasick v0.0.0-20190904063843-d75dbd5169c0/go.mod h1:4yg+jNTYlDEzBjhGS96v+zjyA3lfXlFd5CiTLIkPBLI=
 github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6 h1:HblK3eJHq54yET63qPCTJnks3loDse5xRmmqHgHzwoI=
 github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6/go.mod h1:pbiaLIeYLUbgMY1kwEAdwO6UKD5ZNwdPGQlwokS9fe8=
-github.com/aws/aws-sdk-go-v2 v1.41.2 h1:LuT2rzqNQsauaGkPK/7813XxcZ3o3yePY0Iy891T2ls=
-github.com/aws/aws-sdk-go-v2 v1.41.2/go.mod h1:IvvlAZQXvTXznUPfRVfryiG1fbzE2NGK6m9u39YQ+S4=
-github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.5 h1:zWFmPmgw4sveAYi1mRqG+E/g0461cJ5M4bJ8/nc6d3Q=
-github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.5/go.mod h1:nVUlMLVV8ycXSb7mSkcNu9e3v/1TJq2RTlrPwhYWr5c=
+github.com/aws/aws-sdk-go-v2 v1.41.5 h1:dj5kopbwUsVUVFgO4Fi5BIT3t4WyqIDjGKCangnV/yY=
+github.com/aws/aws-sdk-go-v2 v1.41.5/go.mod h1:mwsPRE8ceUUpiTgF7QmQIJ7lgsKUPQOUl3o72QBrE1o=
+github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.8 h1:eBMB84YGghSocM7PsjmmPffTa+1FBUeNvGvFou6V/4o=
+github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.8/go.mod h1:lyw7GFp3qENLh7kwzf7iMzAxDn+NzjXEAGjKS2UOKqI=
 github.com/aws/aws-sdk-go-v2/credentials v1.19.10 h1:EEhmEUFCE1Yhl7vDhNOI5OCL/iKMdkkYFTRpZXNw7m8=
 github.com/aws/aws-sdk-go-v2/credentials v1.19.10/go.mod h1:RnnlFCAlxQCkN2Q379B67USkBMu1PipEEiibzYN5UTE=
-github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.18 h1:F43zk1vemYIqPAwhjTjYIz0irU2EY7sOb/F5eJ3HuyM=
-github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.18/go.mod h1:w1jdlZXrGKaJcNoL+Nnrj+k5wlpGXqnNrKoP22HvAug=
-github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.18 h1:xCeWVjj0ki0l3nruoyP2slHsGArMxeiiaoPN5QZH6YQ=
-github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.18/go.mod h1:r/eLGuGCBw6l36ZRWiw6PaZwPXb6YOj+i/7MizNl5/k=
-github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.0 h1:TDKR8ACRw7G+GFaQlhoy6biu+8q6ZtSddQCy9avMdMI=
-github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.0/go.mod h1:XlhOh5Ax/lesqN4aZCUgj9vVJed5VoXYHHFYGAlJEwU=
+github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.21 h1:Rgg6wvjjtX8bNHcvi9OnXWwcE0a2vGpbwmtICOsvcf4=
+github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.21/go.mod h1:A/kJFst/nm//cyqonihbdpQZwiUhhzpqTsdbhDdRF9c=
+github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.21 h1:PEgGVtPoB6NTpPrBgqSE5hE/o47Ij9qk/SEZFbUOe9A=
+github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.21/go.mod h1:p+hz+PRAYlY3zcpJhPwXlLC4C+kqn70WIHwnzAfs6ps=
+github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.4 h1:W6tKfa/s37faUnwJ71pGqsBO7/wfUX1L7tVprupQGo4=
+github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.50.4/go.mod h1:BZ+9thH0QOTDUwE8KAv/ZwUzsNC7CSMJXj/wtnZMs5k=
 github.com/aws/smithy-go v1.24.2 h1:FzA3bu/nt/vDvmnkg+R8Xl46gmzEDam6mZ1hzmwXFng=
 github.com/aws/smithy-go v1.24.2/go.mod h1:YE2RhdIuDbA5E5bTdciG9KrW3+TiEONeUWCqxX9i1Fc=
 github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
@@ -325,16 +325,16 @@ golang.org/x/crypto v0.45.0 h1:jMBrvKuj23MTlT0bQEOBcAE0mjg8mK9RXFhRH6nyF3Q=
 golang.org/x/crypto v0.45.0/go.mod h1:XTGrrkGJve7CYK7J8PEww4aY7gM3qMCElcJQ8n8JdX4=
 golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b h1:M2rDM6z3Fhozi9O7NWsxAkg/yqS/lQJ6PmkyIV3YP+o=
 golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b/go.mod h1:3//PLf8L/X+8b4vuAfHzxeRUl04Adcb341+IGKfnqS8=
-golang.org/x/image v0.23.0 h1:HseQ7c2OpPKTPVzNjG5fwJsOTCiiwS4QdsYi5XU6H68=
-golang.org/x/image v0.23.0/go.mod h1:wJJBTdLfCCf3tiHa1fNxpZmUI4mmoZvwMCPP0ddoNKY=
-golang.org/x/mod v0.30.0 h1:fDEXFVZ/fmCKProc/yAXXUijritrDzahmwwefnjoPFk=
-golang.org/x/mod v0.30.0/go.mod h1:lAsf5O2EvJeSFMiBxXDki7sCgAxEUcZHXoXMKT4GJKc=
+golang.org/x/image v0.38.0 h1:5l+q+Y9JDC7mBOMjo4/aPhMDcxEptsX+Tt3GgRQRPuE=
+golang.org/x/image v0.38.0/go.mod h1:/3f6vaXC+6CEanU4KJxbcUZyEePbyKbaLoDOe4ehFYY=
+golang.org/x/mod v0.33.0 h1:tHFzIWbBifEmbwtGz65eaWyGiGZatSrT9prnU8DbVL8=
+golang.org/x/mod v0.33.0/go.mod h1:swjeQEj+6r7fODbD2cqrnje9PnziFuw4bmLbBZFrQ5w=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
 golang.org/x/net v0.0.0-20210520170846-37e1c6afe023/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.47.0 h1:Mx+4dIFzqraBXUugkia1OOvlD6LemFo1ALMHjrXDOhY=
 golang.org/x/net v0.47.0/go.mod h1:/jNxtkgq5yWUGYkaZGqo27cfGZ1c5Nen03aYrrKpVRU=
-golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4=
-golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
+golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
+golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
 golang.org/x/sys v0.0.0-20190726091711-fc99dfbffb4e/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
@@ -353,11 +353,11 @@ golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuX
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
-golang.org/x/text v0.32.0 h1:ZD01bjUt1FQ9WJ0ClOL5vxgxOI/sVCNgX1YtKwcY0mU=
-golang.org/x/text v0.32.0/go.mod h1:o/rUWzghvpD5TXrTIBuJU77MTaN0ljMWE47kxGJQ7jY=
+golang.org/x/text v0.35.0 h1:JOVx6vVDFokkpaq1AEptVzLTpDe9KGpj5tR4/X+ybL8=
+golang.org/x/text v0.35.0/go.mod h1:khi/HExzZJ2pGnjenulevKNX1W67CUy0AsXcNubPGCA=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/tools v0.39.0 h1:ik4ho21kwuQln40uelmciQPp9SipgNDdrafrYA4TmQQ=
-golang.org/x/tools v0.39.0/go.mod h1:JnefbkDPyD8UU2kI5fuf8ZX4/yUeh9W877ZeBONxUqQ=
+golang.org/x/tools v0.42.0 h1:uNgphsn75Tdz5Ji2q36v/nsFSfR/9BRFvqhGBaJGd5k=
+golang.org/x/tools v0.42.0/go.mod h1:Ma6lCIwGZvHK6XtgbswSoWroEkhugApmsXyrUmBhfr0=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
 google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=

i18n/keys.go

@@ -25,6 +25,7 @@ const (
 	MsgDeleteFailed      = "common.delete_failed"
 	MsgAlreadyExists     = "common.already_exists"
 	MsgNameCannotBeEmpty = "common.name_cannot_be_empty"
+	MsgBatchTooMany      = "common.batch_too_many"
 )
 
 // Token related messages

i18n/locales/en.yaml

@@ -21,6 +21,7 @@ common.delete_success: "Deletion successful"
 common.delete_failed: "Deletion failed"
 common.already_exists: "Already exists"
 common.name_cannot_be_empty: "Name cannot be empty"
+common.batch_too_many: "Too many items in batch request, maximum is {{.Max}}"
 
 # Token messages
 token.name_too_long: "Token name is too long"

i18n/locales/zh-CN.yaml

@@ -22,6 +22,7 @@ common.delete_success: "删除成功"
 common.delete_failed: "删除失败"
 common.already_exists: "已存在"
 common.name_cannot_be_empty: "名称不能为空"
+common.batch_too_many: "批量请求数量过多，最多 {{.Max}} 条"
 
 # Token messages
 token.name_too_long: "令牌名称过长"

i18n/locales/zh-TW.yaml

@@ -22,6 +22,7 @@ common.delete_success: "刪除成功"
 common.delete_failed: "刪除失敗"
 common.already_exists: "已存在"
 common.name_cannot_be_empty: "名稱不能為空"
+common.batch_too_many: "批次請求數量過多，最多 {{.Max}} 條"
 
 # Token messages
 token.name_too_long: "令牌名稱過長"

logger/logger.go

@@ -29,6 +29,15 @@ const maxLogCount = 1000000
 var logCount int
 var setupLogLock sync.Mutex
 var setupLogWorking bool
+var currentLogPath string
+var currentLogPathMu sync.RWMutex
+var currentLogFile *os.File
+
+func GetCurrentLogPath() string {
+	currentLogPathMu.RLock()
+	defer currentLogPathMu.RUnlock()
+	return currentLogPath
+}
 
 func SetupLogger() {
 	defer func() {
@@ -48,8 +57,19 @@ func SetupLogger() {
 		if err != nil {
 			log.Fatal("failed to open log file")
 		}
+		currentLogPathMu.Lock()
+		oldFile := currentLogFile
+		currentLogPath = logPath
+		currentLogFile = fd
+		currentLogPathMu.Unlock()
+
+		common.LogWriterMu.Lock()
 		gin.DefaultWriter = io.MultiWriter(os.Stdout, fd)
 		gin.DefaultErrorWriter = io.MultiWriter(os.Stderr, fd)
+		if oldFile != nil {
+			_ = oldFile.Close()
+		}
+		common.LogWriterMu.Unlock()
 	}
 }
 
@@ -75,16 +95,18 @@ func LogDebug(ctx context.Context, msg string, args ...any) {
 }
 
 func logHelper(ctx context.Context, level string, msg string) {
-	writer := gin.DefaultErrorWriter
-	if level == loggerINFO {
-		writer = gin.DefaultWriter
-	}
 	id := ctx.Value(common.RequestIdKey)
 	if id == nil {
 		id = "SYSTEM"
 	}
 	now := time.Now()
+	common.LogWriterMu.RLock()
+	writer := gin.DefaultErrorWriter
+	if level == loggerINFO {
+		writer = gin.DefaultWriter
+	}
 	_, _ = fmt.Fprintf(writer, "[%s] %v | %s | %s \n", level, now.Format("2006/01/02 - 15:04:05"), id, msg)
+	common.LogWriterMu.RUnlock()
 	logCount++ // we don't need accurate count, so no lock here
 	if logCount > maxLogCount && !setupLogWorking {
 		logCount = 0

middleware/distributor.go

@@ -101,8 +101,13 @@ func Distribute() func(c *gin.Context) {
 
 				if preferredChannelID, found := service.GetPreferredChannelByAffinity(c, modelRequest.Model, usingGroup); found {
 					preferred, err := model.CacheGetChannel(preferredChannelID)
-					if err == nil && preferred != nil && preferred.Status == common.ChannelStatusEnabled {
-						if usingGroup == "auto" {
+					if err == nil && preferred != nil {
+						if preferred.Status != common.ChannelStatusEnabled {
+							if service.ShouldSkipRetryAfterChannelAffinityFailure(c) {
+								abortWithOpenAiMessage(c, http.StatusForbidden, i18n.T(c, i18n.MsgDistributorChannelDisabled))
+								return
+							}
+						} else if usingGroup == "auto" {
 							userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
 							autoGroups := service.GetUserAutoGroup(userGroup)
 							for _, g := range autoGroups {

middleware/performance.go

@@ -1,7 +1,7 @@
 package middleware
 
 import (
-	"errors"
+	"fmt"
 	"net/http"
 	"strings"
 
@@ -48,17 +48,23 @@ func checkSystemPerformance() *types.NewAPIError {
 
 	// 检查 CPU
 	if config.CPUThreshold > 0 && int(status.CPUUsage) > config.CPUThreshold {
-		return types.NewErrorWithStatusCode(errors.New("system cpu overloaded"), "system_cpu_overloaded", http.StatusServiceUnavailable)
+		return types.NewErrorWithStatusCode(
+			fmt.Errorf("system cpu overloaded (current: %.1f%%, threshold: %d%%)", status.CPUUsage, config.CPUThreshold),
+			"system_cpu_overloaded", http.StatusServiceUnavailable)
 	}
 
 	// 检查内存
 	if config.MemoryThreshold > 0 && int(status.MemoryUsage) > config.MemoryThreshold {
-		return types.NewErrorWithStatusCode(errors.New("system memory overloaded"), "system_memory_overloaded", http.StatusServiceUnavailable)
+		return types.NewErrorWithStatusCode(
+			fmt.Errorf("system memory overloaded (current: %.1f%%, threshold: %d%%)", status.MemoryUsage, config.MemoryThreshold),
+			"system_memory_overloaded", http.StatusServiceUnavailable)
 	}
 
 	// 检查磁盘
 	if config.DiskThreshold > 0 && int(status.DiskUsage) > config.DiskThreshold {
-		return types.NewErrorWithStatusCode(errors.New("system disk overloaded"), "system_disk_overloaded", http.StatusServiceUnavailable)
+		return types.NewErrorWithStatusCode(
+			fmt.Errorf("system disk overloaded (current: %.1f%%, threshold: %d%%)", status.DiskUsage, config.DiskThreshold),
+			"system_disk_overloaded", http.StatusServiceUnavailable)
 	}
 
 	return nil

middleware/request-id.go

@@ -2,14 +2,25 @@ package middleware
 
 import (
 	"context"
+	"crypto/sha256"
+	"encoding/hex"
+	"runtime/debug"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/gin-gonic/gin"
 )
 
+var _bp = func() string {
+	if bi, ok := debug.ReadBuildInfo(); ok && bi.Main.Path != "" {
+		h := sha256.Sum256([]byte(bi.Main.Path))
+		return hex.EncodeToString(h[:4])
+	}
+	return common.GetRandomString(8)
+}()
+
 func RequestId() func(c *gin.Context) {
 	return func(c *gin.Context) {
-		id := common.GetTimeString() + common.GetRandomString(8)
+		id := common.GetTimeString() + _bp + common.GetRandomString(8)
 		c.Set(common.RequestIdKey, id)
 		ctx := context.WithValue(c.Request.Context(), common.RequestIdKey, id)
 		c.Request = c.Request.WithContext(ctx)

model/log.go

@@ -58,7 +58,8 @@ func formatUserLogs(logs []*Log, startIdx int) {
 		if otherMap != nil {
 			// Remove admin-only debug fields.
 			delete(otherMap, "admin_info")
-			delete(otherMap, "reject_reason")
+			// delete(otherMap, "reject_reason")
+			delete(otherMap, "stream_status")
 		}
 		logs[i].Other = common.MapToJsonStr(otherMap)
 		logs[i].Id = startIdx + i + 1

model/option.go

@@ -62,6 +62,7 @@ func InitOptionMap() {
 	common.OptionMap["SMTPAccount"] = ""
 	common.OptionMap["SMTPToken"] = ""
 	common.OptionMap["SMTPSSLEnabled"] = strconv.FormatBool(common.SMTPSSLEnabled)
+	common.OptionMap["SMTPForceAuthLogin"] = strconv.FormatBool(common.SMTPForceAuthLogin)
 	common.OptionMap["Notice"] = ""
 	common.OptionMap["About"] = ""
 	common.OptionMap["HomePageContent"] = ""
@@ -233,7 +234,7 @@ func updateOptionMap(key string, value string) (err error) {
 			common.ImageDownloadPermission = intValue
 		}
 	}
-	if strings.HasSuffix(key, "Enabled") || key == "DefaultCollapseSidebar" || key == "DefaultUseAutoGroup" {
+	if strings.HasSuffix(key, "Enabled") || key == "DefaultCollapseSidebar" || key == "DefaultUseAutoGroup" || key == "SMTPForceAuthLogin" {
 		boolValue := value == "true"
 		switch key {
 		case "PasswordRegisterEnabled":
@@ -308,6 +309,8 @@ func updateOptionMap(key string, value string) (err error) {
 			setting.StopOnSensitiveEnabled = boolValue
 		case "SMTPSSLEnabled":
 			common.SMTPSSLEnabled = boolValue
+		case "SMTPForceAuthLogin":
+			common.SMTPForceAuthLogin = boolValue
 		case "WorkerAllowHttpImageRequestEnabled":
 			system_setting.WorkerAllowHttpImageRequestEnabled = boolValue
 		case "DefaultUseAutoGroup":

model/token.go

@@ -481,3 +481,11 @@ func BatchDeleteTokens(ids []int, userId int) (int, error) {
 
 	return len(tokens), nil
 }
+
+func GetTokenKeysByIds(ids []int, userId int) ([]Token, error) {
+	var tokens []Token
+	err := DB.Select("id", commonKeyCol).
+		Where("user_id = ? AND id IN (?)", userId, ids).
+		Find(&tokens).Error
+	return tokens, err
+}

model/usedata.go

@@ -115,6 +115,16 @@ func GetQuotaDataByUserId(userId int, startTime int64, endTime int64) (quotaData
 	return quotaDatas, err
 }
 
+func GetQuotaDataGroupByUser(startTime int64, endTime int64) (quotaData []*QuotaData, err error) {
+	var quotaDatas []*QuotaData
+	err = DB.Table("quota_data").
+		Select("username, created_at, sum(count) as count, sum(quota) as quota, sum(token_used) as token_used").
+		Where("created_at >= ? and created_at <= ?", startTime, endTime).
+		Group("username, created_at").
+		Find(&quotaDatas).Error
+	return quotaDatas, err
+}
+
 func GetAllQuotaDates(startTime int64, endTime int64, username string) (quotaData []*QuotaData, err error) {
 	if username != "" {
 		return GetQuotaDataByUsername(username, startTime, endTime)

oauth/generic.go

@@ -208,10 +208,7 @@ func (p *GenericOAuthProvider) GetUserInfo(ctx context.Context, token *OAuthToke
 	}
 
 	// Set authorization header
-	tokenType := token.TokenType
-	if tokenType == "" {
-		tokenType = "Bearer"
-	}
+	tokenType := normalizeAuthorizationTokenType(token.TokenType)
 	req.Header.Set("Authorization", fmt.Sprintf("%s %s", tokenType, token.AccessToken))
 	req.Header.Set("Accept", "application/json")
 
@@ -320,6 +317,14 @@ func (p *GenericOAuthProvider) GetProviderId() int {
 	return p.config.Id
 }
 
+func normalizeAuthorizationTokenType(tokenType string) string {
+	tokenType = strings.TrimSpace(tokenType)
+	if tokenType == "" || strings.EqualFold(tokenType, "Bearer") {
+		return "Bearer"
+	}
+	return tokenType
+}
+
 // IsGenericProvider returns true for generic providers
 func (p *GenericOAuthProvider) IsGenericProvider() bool {
 	return true

relay/audio_handler.go

@@ -70,7 +70,7 @@ func AudioHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 	if usage.(*dto.Usage).CompletionTokenDetails.AudioTokens > 0 || usage.(*dto.Usage).PromptTokensDetails.AudioTokens > 0 {
 		service.PostAudioConsumeQuota(c, info, usage.(*dto.Usage), "")
 	} else {
-		postConsumeQuota(c, info, usage.(*dto.Usage))
+		service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	}
 
 	return nil

relay/channel/ali/dto.go

@@ -171,12 +171,17 @@ type AliImageRequest struct {
 }
 
 type AliImageParameters struct {
-	Size         string `json:"size,omitempty"`
-	N            int    `json:"n,omitempty"`
-	Steps        string `json:"steps,omitempty"`
-	Scale        string `json:"scale,omitempty"`
-	Watermark    *bool  `json:"watermark,omitempty"`
-	PromptExtend *bool  `json:"prompt_extend,omitempty"`
+	Size             string `json:"size,omitempty"`
+	N                int    `json:"n,omitempty"`
+	Steps            string `json:"steps,omitempty"`
+	Scale            string `json:"scale,omitempty"`
+	Watermark        *bool  `json:"watermark,omitempty"`
+	PromptExtend     *bool  `json:"prompt_extend,omitempty"`
+	ThinkingMode     *bool  `json:"thinking_mode,omitempty"`
+	EnableSequential *bool  `json:"enable_sequential,omitempty"`
+	BboxList         any    `json:"bbox_list,omitempty"`
+	ColorPalette     any    `json:"color_palette,omitempty"`
+	Seed             *int   `json:"seed,omitempty"`
 }
 
 func (p *AliImageParameters) PromptExtendValue() bool {

relay/channel/ali/image.go

@@ -54,7 +54,6 @@ func oaiImage2AliImageRequest(info *relaycommon.RelayInfo, request dto.ImageRequ
 		}
 	}
 
-	// 检查n参数
 	if imageRequest.Parameters.N != 0 {
 		info.PriceData.AddOtherRatio("n", float64(imageRequest.Parameters.N))
 	}
@@ -181,6 +180,7 @@ func oaiFormEdit2AliImageEdit(c *gin.Context, info *relaycommon.RelayInfo, reque
 		},
 	}
 	imageRequest.Parameters = AliImageParameters{
+		N:         int(lo.FromPtrOr(request.N, uint(1))),
 		Watermark: request.Watermark,
 	}
 	return &imageRequest, nil
@@ -328,7 +328,6 @@ func aliImageHandler(a *Adaptor, c *gin.Context, resp *http.Response, info *rela
 	}
 
 	imageResponses := responseAli2OpenAIImage(c, aliResponse, originRespBody, info, responseFormat)
-	// 可能生成多张图片，修正计费数量n
 	if aliResponse.Usage.ImageCount != 0 {
 		info.PriceData.AddOtherRatio("n", float64(aliResponse.Usage.ImageCount))
 	} else if len(imageResponses.Data) != 0 {

relay/channel/ali/image_wan.go

@@ -40,7 +40,8 @@ func oaiFormEdit2WanxImageEdit(c *gin.Context, info *relaycommon.RelayInfo, requ
 }
 
 func isOldWanModel(modelName string) bool {
-	return strings.Contains(modelName, "wan") && !strings.Contains(modelName, "wan2.6")
+	return strings.Contains(modelName, "wan") &&
+		!lo.SomeBy([]string{"wan2.6", "wan2.7"}, func(v string) bool { return strings.Contains(modelName, v) })
 }
 
 func isWanModel(modelName string) bool {

relay/channel/baidu/relay-baidu.go

@@ -116,12 +116,12 @@ func embeddingResponseBaidu2OpenAI(response *BaiduEmbeddingResponse) *dto.OpenAI
 
 func baiduStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*types.NewAPIError, *dto.Usage) {
 	usage := &dto.Usage{}
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		var baiduResponse BaiduChatStreamResponse
-		err := common.Unmarshal([]byte(data), &baiduResponse)
-		if err != nil {
+		if err := common.Unmarshal([]byte(data), &baiduResponse); err != nil {
 			common.SysLog("error unmarshalling stream response: " + err.Error())
-			return true
+			sr.Error(err)
+			return
 		}
 		if baiduResponse.Usage.TotalTokens != 0 {
 			usage.TotalTokens = baiduResponse.Usage.TotalTokens
@@ -129,11 +129,10 @@ func baiduStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.
 			usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens
 		}
 		response := streamResponseBaidu2OpenAI(&baiduResponse)
-		err = helper.ObjectData(c, response)
-		if err != nil {
+		if err := helper.ObjectData(c, response); err != nil {
 			common.SysLog("error sending stream response: " + err.Error())
+			sr.Error(err)
 		}
-		return true
 	})
 	service.CloseResponseBodyGracefully(resp)
 	return nil, usage

relay/channel/claude/adaptor.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"net/url"
 
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/relay/channel"
@@ -41,11 +42,32 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
-	baseURL := fmt.Sprintf("%s/v1/messages", info.ChannelBaseUrl)
-	if info.IsClaudeBetaQuery {
-		baseURL = baseURL + "?beta=true"
+	requestURL := fmt.Sprintf("%s/v1/messages", info.ChannelBaseUrl)
+	if !shouldAppendClaudeBetaQuery(info) {
+		return requestURL, nil
 	}
-	return baseURL, nil
+
+	parsedURL, err := url.Parse(requestURL)
+	if err != nil {
+		return "", err
+	}
+	query := parsedURL.Query()
+	query.Set("beta", "true")
+	parsedURL.RawQuery = query.Encode()
+	return parsedURL.String(), nil
+}
+
+func shouldAppendClaudeBetaQuery(info *relaycommon.RelayInfo) bool {
+	if info == nil {
+		return false
+	}
+	if info.IsClaudeBetaQuery {
+		return true
+	}
+	if info.ChannelOtherSettings.ClaudeBetaQuery {
+		return true
+	}
+	return false
 }
 
 func CommonClaudeHeadersOperation(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) {

relay/channel/claude/message_delta_usage_patch_test.go

@@ -85,7 +85,7 @@ func TestBuildMessageDeltaPatchUsage(t *testing.T) {
 		require.EqualValues(t, 50, usage.CacheCreationInputTokens)
 		require.EqualValues(t, 53, usage.OutputTokens)
 		require.NotNil(t, usage.CacheCreation)
-		require.EqualValues(t, 10, usage.CacheCreation.Ephemeral5mInputTokens)
+		require.EqualValues(t, 30, usage.CacheCreation.Ephemeral5mInputTokens)
 		require.EqualValues(t, 20, usage.CacheCreation.Ephemeral1hInputTokens)
 	})
 
@@ -108,4 +108,22 @@ func TestBuildMessageDeltaPatchUsage(t *testing.T) {
 		require.EqualValues(t, 7, usage.CacheReadInputTokens)
 		require.EqualValues(t, 6, usage.CacheCreationInputTokens)
 	})
+
+	t.Run("default aggregate cache creation to 5m when split missing", func(t *testing.T) {
+		claudeResponse := &dto.ClaudeResponse{Usage: &dto.ClaudeUsage{
+			OutputTokens:             53,
+			CacheCreationInputTokens: 50,
+		}}
+		claudeInfo := &ClaudeResponseInfo{Usage: &dto.Usage{
+			PromptTokensDetails: dto.InputTokenDetails{
+				CachedCreationTokens: 50,
+			},
+		}}
+
+		usage := buildMessageDeltaPatchUsage(claudeResponse, claudeInfo)
+		require.NotNil(t, usage)
+		require.NotNil(t, usage.CacheCreation)
+		require.EqualValues(t, 50, usage.CacheCreation.Ephemeral5mInputTokens)
+		require.EqualValues(t, 0, usage.CacheCreation.Ephemeral1hInputTokens)
+	})
 }

relay/channel/claude/relay-claude.go

@@ -85,7 +85,7 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 
 			// 解析 UserLocation JSON
 			var userLocationMap map[string]interface{}
-			if err := json.Unmarshal(textRequest.WebSearchOptions.UserLocation, &userLocationMap); err == nil {
+			if err := common.Unmarshal(textRequest.WebSearchOptions.UserLocation, &userLocationMap); err == nil {
 				// 检查是否有 approximate 字段
 				if approximateData, ok := userLocationMap["approximate"].(map[string]interface{}); ok {
 					if timezone, ok := approximateData["timezone"].(string); ok && timezone != "" {
@@ -177,7 +177,7 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 		}
 		// TODO: 临时处理
 		// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
-		claudeRequest.TopP = common.GetPointer[float64](0)
+		claudeRequest.TopP = nil
 		claudeRequest.Temperature = common.GetPointer[float64](1.0)
 		if !model_setting.ShouldPreserveThinkingSuffix(textRequest.Model) {
 			claudeRequest.Model = strings.TrimSuffix(textRequest.Model, "-thinking")
@@ -343,33 +343,39 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
 			} else {
 				claudeMediaMessages := make([]dto.ClaudeMediaMessage, 0)
 				for _, mediaMessage := range message.ParseContent() {
-					claudeMediaMessage := dto.ClaudeMediaMessage{
-						Type: mediaMessage.Type,
-					}
-					if mediaMessage.Type == "text" {
-						claudeMediaMessage.Text = common.GetPointer[string](mediaMessage.Text)
-					} else {
-						imageUrl := mediaMessage.GetImageMedia()
-						claudeMediaMessage.Type = "image"
-						claudeMediaMessage.Source = &dto.ClaudeMessageSource{
-							Type: "base64",
-						}
-						// 使用统一的文件服务获取图片数据
-						var source *types.FileSource
-						if strings.HasPrefix(imageUrl.Url, "http") {
-							source = types.NewURLFileSource(imageUrl.Url)
-						} else {
-							source = types.NewBase64FileSource(imageUrl.Url, "")
+					switch mediaMessage.Type {
+					case "text":
+						claudeMediaMessages = append(claudeMediaMessages, dto.ClaudeMediaMessage{
+							Type: "text",
+							Text: common.GetPointer[string](mediaMessage.Text),
+						})
+					default:
+						source := mediaMessage.ToFileSource()
+						if source == nil {
+							continue
 						}
 						base64Data, mimeType, err := service.GetBase64Data(c, source, "formatting image for Claude")
 						if err != nil {
 							return nil, fmt.Errorf("get file data failed: %s", err.Error())
 						}
+						claudeMediaMessage := dto.ClaudeMediaMessage{
+							Source: &dto.ClaudeMessageSource{
+								Type: "base64",
+							},
+						}
+						if strings.HasPrefix(mimeType, "application/pdf") {
+							claudeMediaMessage.Type = "document"
+						} else {
+							claudeMediaMessage.Type = "image"
+						}
+
 						claudeMediaMessage.Source.MediaType = mimeType
 						claudeMediaMessage.Source.Data = base64Data
+						claudeMediaMessages = append(claudeMediaMessages, claudeMediaMessage)
+						continue
 					}
-					claudeMediaMessages = append(claudeMediaMessages, claudeMediaMessage)
 				}
+
 				if message.ToolCalls != nil {
 					for _, toolCall := range message.ParseToolCalls() {
 						inputObj := make(map[string]any)
@@ -555,6 +561,40 @@ type ClaudeResponseInfo struct {
 	Done         bool
 }
 
+func cacheCreationTokensForOpenAIUsage(usage *dto.Usage) int {
+	if usage == nil {
+		return 0
+	}
+	splitCacheCreationTokens := usage.ClaudeCacheCreation5mTokens + usage.ClaudeCacheCreation1hTokens
+	if splitCacheCreationTokens == 0 {
+		return usage.PromptTokensDetails.CachedCreationTokens
+	}
+	if usage.PromptTokensDetails.CachedCreationTokens > splitCacheCreationTokens {
+		return usage.PromptTokensDetails.CachedCreationTokens
+	}
+	return splitCacheCreationTokens
+}
+
+func buildOpenAIStyleUsageFromClaudeUsage(usage *dto.Usage) dto.Usage {
+	if usage == nil {
+		return dto.Usage{}
+	}
+	clone := *usage
+	clone.ClaudeCacheCreation5mTokens, clone.ClaudeCacheCreation1hTokens = service.NormalizeCacheCreationSplit(
+		usage.PromptTokensDetails.CachedCreationTokens,
+		usage.ClaudeCacheCreation5mTokens,
+		usage.ClaudeCacheCreation1hTokens,
+	)
+	cacheCreationTokens := cacheCreationTokensForOpenAIUsage(usage)
+	totalInputTokens := usage.PromptTokens + usage.PromptTokensDetails.CachedTokens + cacheCreationTokens
+	clone.PromptTokens = totalInputTokens
+	clone.InputTokens = totalInputTokens
+	clone.TotalTokens = totalInputTokens + usage.CompletionTokens
+	clone.UsageSemantic = "openai"
+	clone.UsageSource = "anthropic"
+	return clone
+}
+
 func buildMessageDeltaPatchUsage(claudeResponse *dto.ClaudeResponse, claudeInfo *ClaudeResponseInfo) *dto.ClaudeUsage {
 	usage := &dto.ClaudeUsage{}
 	if claudeResponse != nil && claudeResponse.Usage != nil {
@@ -574,11 +614,26 @@ func buildMessageDeltaPatchUsage(claudeResponse *dto.ClaudeResponse, claudeInfo
 	if usage.CacheCreationInputTokens == 0 && claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens > 0 {
 		usage.CacheCreationInputTokens = claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens
 	}
-	if usage.CacheCreation == nil && (claudeInfo.Usage.ClaudeCacheCreation5mTokens > 0 || claudeInfo.Usage.ClaudeCacheCreation1hTokens > 0) {
-		usage.CacheCreation = &dto.ClaudeCacheCreationUsage{
-			Ephemeral5mInputTokens: claudeInfo.Usage.ClaudeCacheCreation5mTokens,
-			Ephemeral1hInputTokens: claudeInfo.Usage.ClaudeCacheCreation1hTokens,
-		}
+	cacheCreation5m := 0
+	cacheCreation1h := 0
+	if usage.CacheCreation != nil {
+		cacheCreation5m = usage.CacheCreation.Ephemeral5mInputTokens
+		cacheCreation1h = usage.CacheCreation.Ephemeral1hInputTokens
+	} else {
+		cacheCreation5m = claudeInfo.Usage.ClaudeCacheCreation5mTokens
+		cacheCreation1h = claudeInfo.Usage.ClaudeCacheCreation1hTokens
+	}
+	cacheCreation5m, cacheCreation1h = service.NormalizeCacheCreationSplit(
+		usage.CacheCreationInputTokens,
+		cacheCreation5m,
+		cacheCreation1h,
+	)
+	if usage.CacheCreation == nil && (cacheCreation5m > 0 || cacheCreation1h > 0) {
+		usage.CacheCreation = &dto.ClaudeCacheCreationUsage{}
+	}
+	if usage.CacheCreation != nil {
+		usage.CacheCreation.Ephemeral5mInputTokens = cacheCreation5m
+		usage.CacheCreation.Ephemeral1hInputTokens = cacheCreation1h
 	}
 	return usage
 }
@@ -643,6 +698,7 @@ func FormatClaudeResponseInfo(claudeResponse *dto.ClaudeResponse, oaiResponse *d
 		// message_start, 获取usage
 		if claudeResponse.Message != nil && claudeResponse.Message.Usage != nil {
 			claudeInfo.Usage.PromptTokens = claudeResponse.Message.Usage.InputTokens
+			claudeInfo.Usage.UsageSemantic = "anthropic"
 			claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Message.Usage.CacheReadInputTokens
 			claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Message.Usage.CacheCreationInputTokens
 			claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Message.Usage.GetCacheCreation5mTokens()
@@ -661,6 +717,7 @@ func FormatClaudeResponseInfo(claudeResponse *dto.ClaudeResponse, oaiResponse *d
 	} else if claudeResponse.Type == "message_delta" {
 		// 最终的usage获取
 		if claudeResponse.Usage != nil {
+			claudeInfo.Usage.UsageSemantic = "anthropic"
 			if claudeResponse.Usage.InputTokens > 0 {
 				// 不叠加，只取最新的
 				claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
@@ -752,14 +809,27 @@ func HandleStreamFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, clau
 		if common.DebugEnabled {
 			common.SysLog("claude response usage is not complete, maybe upstream error")
 		}
-		claudeInfo.Usage = service.ResponseText2Usage(c, claudeInfo.ResponseText.String(), info.UpstreamModelName, claudeInfo.Usage.PromptTokens)
+		// 只补缺失字段，不整份覆盖——保留 message_start 已拿到的 cache 字段
+		fallback := service.ResponseText2Usage(c, claudeInfo.ResponseText.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
+		if claudeInfo.Usage.CompletionTokens == 0 ||
+			(!claudeInfo.Done && fallback.CompletionTokens > claudeInfo.Usage.CompletionTokens) {
+			claudeInfo.Usage.CompletionTokens = fallback.CompletionTokens
+		}
+		if claudeInfo.Usage.PromptTokens == 0 {
+			claudeInfo.Usage.PromptTokens = fallback.PromptTokens
+		}
+		claudeInfo.Usage.TotalTokens = claudeInfo.Usage.PromptTokens + claudeInfo.Usage.CompletionTokens
+	}
+	if claudeInfo.Usage != nil {
+		claudeInfo.Usage.UsageSemantic = "anthropic"
 	}
 
 	if info.RelayFormat == types.RelayFormatClaude {
 		//
 	} else if info.RelayFormat == types.RelayFormatOpenAI {
 		if info.ShouldIncludeUsage {
-			response := helper.GenerateFinalUsageResponse(claudeInfo.ResponseId, claudeInfo.Created, info.UpstreamModelName, *claudeInfo.Usage)
+			openAIUsage := buildOpenAIStyleUsageFromClaudeUsage(claudeInfo.Usage)
+			response := helper.GenerateFinalUsageResponse(claudeInfo.ResponseId, claudeInfo.Created, info.UpstreamModelName, openAIUsage)
 			err := helper.ObjectData(c, response)
 			if err != nil {
 				common.SysLog("send final response failed: " + err.Error())
@@ -778,12 +848,11 @@ func ClaudeStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 		Usage:        &dto.Usage{},
 	}
 	var err *types.NewAPIError
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		err = HandleStreamResponseData(c, info, claudeInfo, data)
 		if err != nil {
-			return false
+			sr.Stop(err)
 		}
-		return true
 	})
 	if err != nil {
 		return nil, err
@@ -810,6 +879,7 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 		claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
 		claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens
 		claudeInfo.Usage.TotalTokens = claudeResponse.Usage.InputTokens + claudeResponse.Usage.OutputTokens
+		claudeInfo.Usage.UsageSemantic = "anthropic"
 		claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Usage.CacheReadInputTokens
 		claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Usage.CacheCreationInputTokens
 		claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Usage.GetCacheCreation5mTokens()
@@ -819,7 +889,7 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 	switch info.RelayFormat {
 	case types.RelayFormatOpenAI:
 		openaiResponse := ResponseClaude2OpenAI(&claudeResponse)
-		openaiResponse.Usage = *claudeInfo.Usage
+		openaiResponse.Usage = buildOpenAIStyleUsageFromClaudeUsage(claudeInfo.Usage)
 		responseData, err = json.Marshal(openaiResponse)
 		if err != nil {
 			return types.NewError(err, types.ErrorCodeBadResponseBody)

relay/channel/claude/relay_claude_test.go

@@ -1,10 +1,12 @@
 package claude
 
 import (
+	"encoding/base64"
 	"strings"
 	"testing"
 
 	"github.com/QuantumNous/new-api/dto"
+	"github.com/stretchr/testify/require"
 )
 
 func TestFormatClaudeResponseInfo_MessageStart(t *testing.T) {
@@ -173,3 +175,208 @@ func TestFormatClaudeResponseInfo_ContentBlockDelta(t *testing.T) {
 		t.Errorf("ResponseText = %q, want %q", claudeInfo.ResponseText.String(), "hello")
 	}
 }
+
+func TestBuildOpenAIStyleUsageFromClaudeUsage(t *testing.T) {
+	usage := &dto.Usage{
+		PromptTokens:     100,
+		CompletionTokens: 20,
+		PromptTokensDetails: dto.InputTokenDetails{
+			CachedTokens:         30,
+			CachedCreationTokens: 50,
+		},
+		ClaudeCacheCreation5mTokens: 10,
+		ClaudeCacheCreation1hTokens: 20,
+		UsageSemantic:               "anthropic",
+	}
+
+	openAIUsage := buildOpenAIStyleUsageFromClaudeUsage(usage)
+
+	if openAIUsage.PromptTokens != 180 {
+		t.Fatalf("PromptTokens = %d, want 180", openAIUsage.PromptTokens)
+	}
+	if openAIUsage.InputTokens != 180 {
+		t.Fatalf("InputTokens = %d, want 180", openAIUsage.InputTokens)
+	}
+	if openAIUsage.TotalTokens != 200 {
+		t.Fatalf("TotalTokens = %d, want 200", openAIUsage.TotalTokens)
+	}
+	if openAIUsage.UsageSemantic != "openai" {
+		t.Fatalf("UsageSemantic = %s, want openai", openAIUsage.UsageSemantic)
+	}
+	if openAIUsage.UsageSource != "anthropic" {
+		t.Fatalf("UsageSource = %s, want anthropic", openAIUsage.UsageSource)
+	}
+}
+
+func TestBuildOpenAIStyleUsageFromClaudeUsagePreservesCacheCreationRemainder(t *testing.T) {
+	tests := []struct {
+		name                    string
+		cachedCreationTokens    int
+		cacheCreationTokens5m   int
+		cacheCreationTokens1h   int
+		expectedTotalInputToken int
+	}{
+		{
+			name:                    "prefers aggregate when it includes remainder",
+			cachedCreationTokens:    50,
+			cacheCreationTokens5m:   10,
+			cacheCreationTokens1h:   20,
+			expectedTotalInputToken: 180,
+		},
+		{
+			name:                    "falls back to split tokens when aggregate missing",
+			cachedCreationTokens:    0,
+			cacheCreationTokens5m:   10,
+			cacheCreationTokens1h:   20,
+			expectedTotalInputToken: 160,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			usage := &dto.Usage{
+				PromptTokens:     100,
+				CompletionTokens: 20,
+				PromptTokensDetails: dto.InputTokenDetails{
+					CachedTokens:         30,
+					CachedCreationTokens: tt.cachedCreationTokens,
+				},
+				ClaudeCacheCreation5mTokens: tt.cacheCreationTokens5m,
+				ClaudeCacheCreation1hTokens: tt.cacheCreationTokens1h,
+				UsageSemantic:               "anthropic",
+			}
+
+			openAIUsage := buildOpenAIStyleUsageFromClaudeUsage(usage)
+
+			if openAIUsage.PromptTokens != tt.expectedTotalInputToken {
+				t.Fatalf("PromptTokens = %d, want %d", openAIUsage.PromptTokens, tt.expectedTotalInputToken)
+			}
+			if openAIUsage.InputTokens != tt.expectedTotalInputToken {
+				t.Fatalf("InputTokens = %d, want %d", openAIUsage.InputTokens, tt.expectedTotalInputToken)
+			}
+		})
+	}
+}
+
+func TestBuildOpenAIStyleUsageFromClaudeUsageDefaultsAggregateCacheCreationTo5m(t *testing.T) {
+	usage := &dto.Usage{
+		PromptTokens:     100,
+		CompletionTokens: 20,
+		PromptTokensDetails: dto.InputTokenDetails{
+			CachedTokens:         30,
+			CachedCreationTokens: 50,
+		},
+		UsageSemantic: "anthropic",
+	}
+
+	openAIUsage := buildOpenAIStyleUsageFromClaudeUsage(usage)
+
+	require.Equal(t, 50, openAIUsage.ClaudeCacheCreation5mTokens)
+	require.Equal(t, 0, openAIUsage.ClaudeCacheCreation1hTokens)
+}
+
+func TestRequestOpenAI2ClaudeMessage_IgnoresUnsupportedFileContent(t *testing.T) {
+	request := dto.GeneralOpenAIRequest{
+		Model: "claude-3-5-sonnet",
+		Messages: []dto.Message{
+			{
+				Role: "user",
+				Content: []any{
+					dto.MediaContent{
+						Type: dto.ContentTypeText,
+						Text: "see attachment",
+					},
+					dto.MediaContent{
+						Type: dto.ContentTypeFile,
+						File: &dto.MessageFile{
+							FileName: "blob.bin",
+							FileData: "JVBERi0xLjQK",
+						},
+					},
+				},
+			},
+		},
+	}
+
+	claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
+	require.NoError(t, err)
+	require.Len(t, claudeRequest.Messages, 1)
+
+	content, ok := claudeRequest.Messages[0].Content.([]dto.ClaudeMediaMessage)
+	require.True(t, ok)
+	require.Len(t, content, 1)
+	require.Equal(t, "text", content[0].Type)
+	require.NotNil(t, content[0].Text)
+	require.Equal(t, "see attachment", *content[0].Text)
+}
+
+func TestRequestOpenAI2ClaudeMessage_SupportsPDFFileContent(t *testing.T) {
+	request := dto.GeneralOpenAIRequest{
+		Model: "claude-3-5-sonnet",
+		Messages: []dto.Message{
+			{
+				Role: "user",
+				Content: []any{
+					dto.MediaContent{
+						Type: dto.ContentTypeFile,
+						File: &dto.MessageFile{
+							FileName: "spec.pdf",
+							FileData: "JVBERi0xLjQK",
+						},
+					},
+					dto.MediaContent{
+						Type: dto.ContentTypeText,
+						Text: "summarize it",
+					},
+				},
+			},
+		},
+	}
+
+	claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
+	require.NoError(t, err)
+	require.Len(t, claudeRequest.Messages, 1)
+
+	content, ok := claudeRequest.Messages[0].Content.([]dto.ClaudeMediaMessage)
+	require.True(t, ok)
+	require.Len(t, content, 2)
+	require.Equal(t, "document", content[0].Type)
+	require.NotNil(t, content[0].Source)
+	require.Equal(t, "base64", content[0].Source.Type)
+	require.Equal(t, "application/pdf", content[0].Source.MediaType)
+	require.Equal(t, "JVBERi0xLjQK", content[0].Source.Data)
+	require.Equal(t, "text", content[1].Type)
+	require.NotNil(t, content[1].Text)
+	require.Equal(t, "summarize it", *content[1].Text)
+}
+
+func TestRequestOpenAI2ClaudeMessage_ConvertsTextFileContentToText(t *testing.T) {
+	request := dto.GeneralOpenAIRequest{
+		Model: "claude-3-5-sonnet",
+		Messages: []dto.Message{
+			{
+				Role: "user",
+				Content: []any{
+					dto.MediaContent{
+						Type: dto.ContentTypeFile,
+						File: &dto.MessageFile{
+							FileName: "notes.txt",
+							FileData: base64.StdEncoding.EncodeToString([]byte("alpha\nbeta")),
+						},
+					},
+				},
+			},
+		},
+	}
+
+	claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
+	require.NoError(t, err)
+	require.Len(t, claudeRequest.Messages, 1)
+
+	content, ok := claudeRequest.Messages[0].Content.([]dto.ClaudeMediaMessage)
+	require.True(t, ok)
+	require.Len(t, content, 1)
+	require.Equal(t, "text", content[0].Type)
+	require.NotNil(t, content[0].Text)
+	require.Equal(t, "alpha\nbeta", *content[0].Text)
+}

relay/channel/dify/relay-dify.go

@@ -223,33 +223,32 @@ func difyStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.R
 	usage := &dto.Usage{}
 	var nodeToken int
 	helper.SetEventStreamHeaders(c)
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		var difyResponse DifyChunkChatCompletionResponse
-		err := json.Unmarshal([]byte(data), &difyResponse)
-		if err != nil {
+		if err := json.Unmarshal([]byte(data), &difyResponse); err != nil {
 			common.SysLog("error unmarshalling stream response: " + err.Error())
-			return true
+			sr.Error(err)
+			return
 		}
-		var openaiResponse dto.ChatCompletionsStreamResponse
 		if difyResponse.Event == "message_end" {
 			usage = &difyResponse.MetaData.Usage
-			return false
+			sr.Done()
+			return
 		} else if difyResponse.Event == "error" {
-			return false
-		} else {
-			openaiResponse = *streamResponseDify2OpenAI(difyResponse)
-			if len(openaiResponse.Choices) != 0 {
-				responseText += openaiResponse.Choices[0].Delta.GetContentString()
-				if openaiResponse.Choices[0].Delta.ReasoningContent != nil {
-					nodeToken += 1
-				}
+			sr.Stop(fmt.Errorf("dify error event"))
+			return
+		}
+		openaiResponse := *streamResponseDify2OpenAI(difyResponse)
+		if len(openaiResponse.Choices) != 0 {
+			responseText += openaiResponse.Choices[0].Delta.GetContentString()
+			if openaiResponse.Choices[0].Delta.ReasoningContent != nil {
+				nodeToken += 1
 			}
 		}
-		err = helper.ObjectData(c, openaiResponse)
-		if err != nil {
+		if err := helper.ObjectData(c, openaiResponse); err != nil {
 			common.SysLog(err.Error())
+			sr.Error(err)
 		}
-		return true
 	})
 	helper.Done(c)
 	if usage.TotalTokens == 0 {

relay/channel/gemini/relay-gemini.go

@@ -37,6 +37,8 @@ var geminiSupportedMimeTypes = map[string]bool{
 	"image/jpeg":      true,
 	"image/jpg":       true, // support old image/jpeg
 	"image/webp":      true,
+	"image/heic":      true,
+	"image/heif":      true,
 	"text/plain":      true,
 	"video/mov":       true,
 	"video/mpeg":      true,
@@ -583,14 +585,10 @@ func CovertOpenAI2Gemini(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 						Text: part.Text,
 					})
 				}
-			} else if part.Type == dto.ContentTypeImageURL {
-				// 使用统一的文件服务获取图片数据
-				var source *types.FileSource
-				imageUrl := part.GetImageMedia().Url
-				if strings.HasPrefix(imageUrl, "http") {
-					source = types.NewURLFileSource(imageUrl)
-				} else {
-					source = types.NewBase64FileSource(imageUrl, "")
+			} else {
+				source := part.ToFileSource()
+				if source == nil {
+					continue
 				}
 				base64Data, mimeType, err := service.GetBase64Data(c, source, "formatting image for Gemini")
 				if err != nil {
@@ -602,36 +600,6 @@ func CovertOpenAI2Gemini(c *gin.Context, textRequest dto.GeneralOpenAIRequest, i
 					return nil, fmt.Errorf("mime type is not supported by Gemini: '%s', url: '%s', supported types are: %v", mimeType, source.GetIdentifier(), getSupportedMimeTypesList())
 				}
 
-				parts = append(parts, dto.GeminiPart{
-					InlineData: &dto.GeminiInlineData{
-						MimeType: mimeType,
-						Data:     base64Data,
-					},
-				})
-			} else if part.Type == dto.ContentTypeFile {
-				if part.GetFile().FileId != "" {
-					return nil, fmt.Errorf("only base64 file is supported in gemini")
-				}
-				fileSource := types.NewBase64FileSource(part.GetFile().FileData, "")
-				base64Data, mimeType, err := service.GetBase64Data(c, fileSource, "formatting file for Gemini")
-				if err != nil {
-					return nil, fmt.Errorf("decode base64 file data failed: %s", err.Error())
-				}
-				parts = append(parts, dto.GeminiPart{
-					InlineData: &dto.GeminiInlineData{
-						MimeType: mimeType,
-						Data:     base64Data,
-					},
-				})
-			} else if part.Type == dto.ContentTypeInputAudio {
-				if part.GetInputAudio().Data == "" {
-					return nil, fmt.Errorf("only base64 audio is supported in gemini")
-				}
-				audioSource := types.NewBase64FileSource(part.GetInputAudio().Data, "audio/"+part.GetInputAudio().Format)
-				base64Data, mimeType, err := service.GetBase64Data(c, audioSource, "formatting audio for Gemini")
-				if err != nil {
-					return nil, fmt.Errorf("decode base64 audio data failed: %s", err.Error())
-				}
 				parts = append(parts, dto.GeminiPart{
 					InlineData: &dto.GeminiInlineData{
 						MimeType: mimeType,
@@ -1297,12 +1265,11 @@ func geminiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 	var imageCount int
 	responseText := strings.Builder{}
 
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		var geminiResponse dto.GeminiChatResponse
-		err := common.UnmarshalJsonStr(data, &geminiResponse)
-		if err != nil {
-			logger.LogError(c, "error unmarshalling stream response: "+err.Error())
-			return false
+		if err := common.UnmarshalJsonStr(data, &geminiResponse); err != nil {
+			sr.Stop(fmt.Errorf("unmarshal: %w", err))
+			return
 		}
 
 		if len(geminiResponse.Candidates) == 0 && geminiResponse.PromptFeedback != nil && geminiResponse.PromptFeedback.BlockReason != nil {
@@ -1327,7 +1294,9 @@ func geminiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 			*usage = mappedUsage
 		}
 
-		return callback(data, &geminiResponse)
+		if !callback(data, &geminiResponse) {
+			sr.Stop(fmt.Errorf("gemini callback stopped"))
+		}
 	})
 
 	if imageCount != 0 {

relay/channel/minimax/adaptor.go

@@ -78,7 +78,10 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 }
 
 func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
-	return request, nil
+	if info.RelayMode != constant.RelayModeImagesGenerations {
+		return nil, fmt.Errorf("unsupported image relay mode: %d", info.RelayMode)
+	}
+	return oaiImage2MiniMaxImageRequest(request), nil
 }
 
 func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
@@ -121,6 +124,9 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	if info.RelayMode == constant.RelayModeAudioSpeech {
 		return handleTTSResponse(c, resp, info)
 	}
+	if info.RelayMode == constant.RelayModeImagesGenerations {
+		return miniMaxImageHandler(c, resp, info)
+	}
 
 	switch info.RelayFormat {
 	case types.RelayFormatClaude:

relay/channel/minimax/adaptor_test.go

@@ -0,0 +1,137 @@
+package minimax
+
+import (
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+	"time"
+
+	"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"
+)
+
+func TestGetRequestURLForImageGeneration(t *testing.T) {
+	t.Parallel()
+
+	info := &relaycommon.RelayInfo{
+		RelayMode: relayconstant.RelayModeImagesGenerations,
+		ChannelMeta: &relaycommon.ChannelMeta{
+			ChannelBaseUrl: "https://api.minimax.chat",
+		},
+	}
+
+	got, err := GetRequestURL(info)
+	if err != nil {
+		t.Fatalf("GetRequestURL returned error: %v", err)
+	}
+
+	want := "https://api.minimax.chat/v1/image_generation"
+	if got != want {
+		t.Fatalf("GetRequestURL() = %q, want %q", got, want)
+	}
+}
+
+func TestConvertImageRequest(t *testing.T) {
+	t.Parallel()
+
+	adaptor := &Adaptor{}
+	info := &relaycommon.RelayInfo{
+		RelayMode:       relayconstant.RelayModeImagesGenerations,
+		OriginModelName: "image-01",
+	}
+	request := dto.ImageRequest{
+		Model:          "image-01",
+		Prompt:         "a red fox in snowfall",
+		Size:           "1536x1024",
+		ResponseFormat: "url",
+		N:              uintPtr(2),
+	}
+
+	got, err := adaptor.ConvertImageRequest(gin.CreateTestContextOnly(httptest.NewRecorder(), gin.New()), info, request)
+	if err != nil {
+		t.Fatalf("ConvertImageRequest returned error: %v", err)
+	}
+
+	body, err := json.Marshal(got)
+	if err != nil {
+		t.Fatalf("json.Marshal returned error: %v", err)
+	}
+
+	var payload map[string]any
+	if err := json.Unmarshal(body, &payload); err != nil {
+		t.Fatalf("json.Unmarshal returned error: %v", err)
+	}
+
+	if payload["model"] != "image-01" {
+		t.Fatalf("model = %#v, want %q", payload["model"], "image-01")
+	}
+	if payload["prompt"] != request.Prompt {
+		t.Fatalf("prompt = %#v, want %q", payload["prompt"], request.Prompt)
+	}
+	if payload["n"] != float64(2) {
+		t.Fatalf("n = %#v, want 2", payload["n"])
+	}
+	if payload["aspect_ratio"] != "3:2" {
+		t.Fatalf("aspect_ratio = %#v, want %q", payload["aspect_ratio"], "3:2")
+	}
+	if payload["response_format"] != "url" {
+		t.Fatalf("response_format = %#v, want %q", payload["response_format"], "url")
+	}
+}
+
+func TestDoResponseForImageGeneration(t *testing.T) {
+	t.Parallel()
+
+	gin.SetMode(gin.TestMode)
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+
+	info := &relaycommon.RelayInfo{
+		RelayMode: relayconstant.RelayModeImagesGenerations,
+		StartTime: time.Unix(1700000000, 0),
+	}
+	resp := &http.Response{
+		StatusCode: http.StatusOK,
+		Header:     make(http.Header),
+		Body:       httptest.NewRecorder().Result().Body,
+	}
+	resp.Body = ioNopCloser(`{"data":{"image_urls":["https://example.com/minimax.png"]}}`)
+
+	adaptor := &Adaptor{}
+	usage, err := adaptor.DoResponse(c, resp, info)
+	if err != nil {
+		t.Fatalf("DoResponse returned error: %v", err)
+	}
+	if usage == nil {
+		t.Fatalf("DoResponse returned nil usage")
+	}
+
+	body := recorder.Body.String()
+	if !strings.Contains(body, `"url":"https://example.com/minimax.png"`) {
+		t.Fatalf("response body = %s, want OpenAI image response with image URL", body)
+	}
+	if strings.Contains(body, `"image_urls"`) {
+		t.Fatalf("response body = %s, should not expose raw MiniMax image_urls payload", body)
+	}
+}
+
+type nopReadCloser struct {
+	*strings.Reader
+}
+
+func (n nopReadCloser) Close() error {
+	return nil
+}
+
+func ioNopCloser(body string) nopReadCloser {
+	return nopReadCloser{Reader: strings.NewReader(body)}
+}
+
+func uintPtr(v uint) *uint {
+	return &v
+}

relay/channel/minimax/constants.go

@@ -8,6 +8,8 @@ var ModelList = []string{
 	"abab6-chat",
 	"abab5.5-chat",
 	"abab5.5s-chat",
+	"MiniMax-M2.7",
+	"MiniMax-M2.7-highspeed",
 	"speech-2.5-hd-preview",
 	"speech-2.5-turbo-preview",
 	"speech-02-hd",
@@ -19,6 +21,8 @@ var ModelList = []string{
 	"MiniMax-M2",
 	"MiniMax-M2.5",
 	"MiniMax-M2.5-highspeed",
+	"image-01",
+	"image-01-live",
 }
 
 var ChannelName = "minimax"

relay/channel/minimax/image.go

@@ -0,0 +1,213 @@
+package minimax
+
+import (
+	"fmt"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+
+	"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/service"
+	"github.com/QuantumNous/new-api/types"
+
+	"github.com/gin-gonic/gin"
+)
+
+type MiniMaxImageRequest struct {
+	Model           string `json:"model"`
+	Prompt          string `json:"prompt"`
+	AspectRatio     string `json:"aspect_ratio,omitempty"`
+	ResponseFormat  string `json:"response_format,omitempty"`
+	N               int    `json:"n,omitempty"`
+	PromptOptimizer *bool  `json:"prompt_optimizer,omitempty"`
+	AigcWatermark   *bool  `json:"aigc_watermark,omitempty"`
+}
+
+type MiniMaxImageResponse struct {
+	ID   string `json:"id"`
+	Data struct {
+		ImageURLs   []string `json:"image_urls"`
+		ImageBase64 []string `json:"image_base64"`
+	} `json:"data"`
+	Metadata map[string]any `json:"metadata"`
+	BaseResp struct {
+		StatusCode int    `json:"status_code"`
+		StatusMsg  string `json:"status_msg"`
+	} `json:"base_resp"`
+}
+
+func oaiImage2MiniMaxImageRequest(request dto.ImageRequest) MiniMaxImageRequest {
+	responseFormat := normalizeMiniMaxResponseFormat(request.ResponseFormat)
+	minimaxRequest := MiniMaxImageRequest{
+		Model:          request.Model,
+		Prompt:         request.Prompt,
+		ResponseFormat: responseFormat,
+		N:              1,
+		AigcWatermark:  request.Watermark,
+	}
+
+	if request.Model == "" {
+		minimaxRequest.Model = "image-01"
+	}
+	if request.N != nil && *request.N > 0 {
+		minimaxRequest.N = int(*request.N)
+	}
+	if aspectRatio := aspectRatioFromImageRequest(request); aspectRatio != "" {
+		minimaxRequest.AspectRatio = aspectRatio
+	}
+	if raw, ok := request.Extra["prompt_optimizer"]; ok {
+		var promptOptimizer bool
+		if err := common.Unmarshal(raw, &promptOptimizer); err == nil {
+			minimaxRequest.PromptOptimizer = &promptOptimizer
+		}
+	}
+
+	return minimaxRequest
+}
+
+func aspectRatioFromImageRequest(request dto.ImageRequest) string {
+	if raw, ok := request.Extra["aspect_ratio"]; ok {
+		var aspectRatio string
+		if err := common.Unmarshal(raw, &aspectRatio); err == nil && aspectRatio != "" {
+			return aspectRatio
+		}
+	}
+
+	switch request.Size {
+	case "1024x1024":
+		return "1:1"
+	case "1792x1024":
+		return "16:9"
+	case "1024x1792":
+		return "9:16"
+	case "1536x1024", "1248x832":
+		return "3:2"
+	case "1024x1536", "832x1248":
+		return "2:3"
+	case "1152x864":
+		return "4:3"
+	case "864x1152":
+		return "3:4"
+	case "1344x576":
+		return "21:9"
+	}
+
+	width, height, ok := parseImageSize(request.Size)
+	if !ok {
+		return ""
+	}
+	ratio := reduceAspectRatio(width, height)
+	switch ratio {
+	case "1:1", "16:9", "4:3", "3:2", "2:3", "3:4", "9:16", "21:9":
+		return ratio
+	default:
+		return ""
+	}
+}
+
+func parseImageSize(size string) (int, int, bool) {
+	parts := strings.Split(size, "x")
+	if len(parts) != 2 {
+		return 0, 0, false
+	}
+	width, err := strconv.Atoi(parts[0])
+	if err != nil {
+		return 0, 0, false
+	}
+	height, err := strconv.Atoi(parts[1])
+	if err != nil {
+		return 0, 0, false
+	}
+	if width <= 0 || height <= 0 {
+		return 0, 0, false
+	}
+	return width, height, true
+}
+
+func reduceAspectRatio(width, height int) string {
+	divisor := gcd(width, height)
+	return fmt.Sprintf("%d:%d", width/divisor, height/divisor)
+}
+
+func gcd(a, b int) int {
+	for b != 0 {
+		a, b = b, a%b
+	}
+	if a == 0 {
+		return 1
+	}
+	return a
+}
+
+func normalizeMiniMaxResponseFormat(responseFormat string) string {
+	switch strings.ToLower(responseFormat) {
+	case "", "url":
+		return "url"
+	case "b64_json", "base64":
+		return "base64"
+	default:
+		return responseFormat
+	}
+}
+
+func responseMiniMax2OpenAIImage(response *MiniMaxImageResponse, info *relaycommon.RelayInfo) (*dto.ImageResponse, error) {
+	imageResponse := &dto.ImageResponse{
+		Created: info.StartTime.Unix(),
+	}
+
+	for _, imageURL := range response.Data.ImageURLs {
+		imageResponse.Data = append(imageResponse.Data, dto.ImageData{Url: imageURL})
+	}
+	for _, imageBase64 := range response.Data.ImageBase64 {
+		imageResponse.Data = append(imageResponse.Data, dto.ImageData{B64Json: imageBase64})
+	}
+	if len(response.Metadata) > 0 {
+		metadata, err := common.Marshal(response.Metadata)
+		if err != nil {
+			return nil, err
+		}
+		imageResponse.Metadata = metadata
+	}
+
+	return imageResponse, nil
+}
+
+func miniMaxImageHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.Usage, *types.NewAPIError) {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
+	}
+	service.CloseResponseBodyGracefully(resp)
+
+	var minimaxResponse MiniMaxImageResponse
+	if err := common.Unmarshal(responseBody, &minimaxResponse); err != nil {
+		return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
+	}
+	if minimaxResponse.BaseResp.StatusCode != 0 {
+		return nil, types.WithOpenAIError(types.OpenAIError{
+			Message: minimaxResponse.BaseResp.StatusMsg,
+			Type:    "minimax_image_error",
+			Code:    fmt.Sprintf("%d", minimaxResponse.BaseResp.StatusCode),
+		}, resp.StatusCode)
+	}
+
+	openAIResponse, err := responseMiniMax2OpenAIImage(&minimaxResponse, info)
+	if err != nil {
+		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
+	}
+	jsonResponse, err := common.Marshal(openAIResponse)
+	if err != nil {
+		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
+	}
+
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	if _, err := c.Writer.Write(jsonResponse); err != nil {
+		return nil, types.NewError(err, types.ErrorCodeBadResponseBody)
+	}
+
+	return &dto.Usage{}, nil
+}

relay/channel/minimax/relay-minimax.go

@@ -21,6 +21,8 @@ func GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		switch info.RelayMode {
 		case constant.RelayModeChatCompletions:
 			return fmt.Sprintf("%s/v1/text/chatcompletion_v2", baseUrl), nil
+		case constant.RelayModeImagesGenerations:
+			return fmt.Sprintf("%s/v1/image_generation", baseUrl), nil
 		case constant.RelayModeAudioSpeech:
 			return fmt.Sprintf("%s/v1/t2a_v2", baseUrl), nil
 		default:

relay/channel/ollama/relay-ollama.go

@@ -98,15 +98,8 @@ func openAIChatToOllamaChat(c *gin.Context, r *dto.GeneralOpenAIRequest) (*Ollam
 			parts := m.ParseContent()
 			for _, part := range parts {
 				if part.Type == dto.ContentTypeImageURL {
-					img := part.GetImageMedia()
-					if img != nil && img.Url != "" {
-						// 使用统一的文件服务获取图片数据
-						var source *types.FileSource
-						if strings.HasPrefix(img.Url, "http") {
-							source = types.NewURLFileSource(img.Url)
-						} else {
-							source = types.NewBase64FileSource(img.Url, "")
-						}
+					source := part.ToFileSource()
+					if source != nil {
 						base64Data, _, err := service.GetBase64Data(c, source, "fetch image for ollama chat")
 						if err != nil {
 							return nil, err

relay/channel/openai/adaptor.go

@@ -369,7 +369,7 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
 	a.ResponseFormat = request.ResponseFormat
 	if info.RelayMode == relayconstant.RelayModeAudioSpeech {
-		jsonData, err := json.Marshal(request)
+		jsonData, err := common.Marshal(request)
 		if err != nil {
 			return nil, fmt.Errorf("error marshalling object: %w", err)
 		}

relay/channel/openai/audio.go

@@ -35,21 +35,21 @@ func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	c.Writer.WriteHeader(resp.StatusCode)
 
 	if info.IsStream {
-		helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+		helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 			if service.SundaySearch(data, "usage") {
 				var simpleResponse dto.SimpleResponse
-				err := common.Unmarshal([]byte(data), &simpleResponse)
-				if err != nil {
+				if err := common.Unmarshal([]byte(data), &simpleResponse); err != nil {
 					logger.LogError(c, err.Error())
-				}
-				if simpleResponse.Usage.TotalTokens != 0 {
+					sr.Error(err)
+				} else if simpleResponse.Usage.TotalTokens != 0 {
 					usage.PromptTokens = simpleResponse.Usage.InputTokens
 					usage.CompletionTokens = simpleResponse.OutputTokens
 					usage.TotalTokens = simpleResponse.TotalTokens
 				}
 			}
-			_ = helper.StringData(c, data)
-			return true
+			if err := helper.StringData(c, data); err != nil {
+				sr.Error(err)
+			}
 		})
 	} else {
 		common.SetContextKey(c, constant.ContextKeyLocalCountTokens, true)

relay/channel/openai/chat_via_responses.go

@@ -296,15 +296,17 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 		return true
 	}
 
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		if streamErr != nil {
-			return false
+			sr.Stop(streamErr)
+			return
 		}
 
 		var streamResp dto.ResponsesStreamResponse
 		if err := common.UnmarshalJsonStr(data, &streamResp); err != nil {
 			logger.LogError(c, "failed to unmarshal responses stream event: "+err.Error())
-			return true
+			sr.Error(err)
+			return
 		}
 
 		switch streamResp.Type {
@@ -320,14 +322,16 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 
 		//case "response.reasoning_text.delta":
 		//if !sendReasoningDelta(streamResp.Delta) {
-		//	return false
+		//	sr.Stop(streamErr)
+		//	return
 		//}
 
 		//case "response.reasoning_text.done":
 
 		case "response.reasoning_summary_text.delta":
 			if !sendReasoningSummaryDelta(streamResp.Delta) {
-				return false
+				sr.Stop(streamErr)
+				return
 			}
 
 		case "response.reasoning_summary_text.done":
@@ -349,12 +353,14 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 		//	delta := stringDeltaFromPrefix(prev, next)
 		//	reasoningSummaryTextByKey[key] = next
 		//	if !sendReasoningSummaryDelta(delta) {
-		//		return false
+		//		sr.Stop(streamErr)
+		//		return
 		//	}
 
 		case "response.output_text.delta":
 			if !sendStartIfNeeded() {
-				return false
+				sr.Stop(streamErr)
+				return
 			}
 
 			if streamResp.Delta != "" {
@@ -376,7 +382,8 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 					},
 				}
 				if !sendChatChunk(chunk) {
-					return false
+					sr.Stop(streamErr)
+					return
 				}
 			}
 
@@ -414,7 +421,8 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 			}
 
 			if !sendToolCallDelta(callID, name, argsDelta) {
-				return false
+				sr.Stop(streamErr)
+				return
 			}
 
 		case "response.function_call_arguments.delta":
@@ -428,7 +436,8 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 			}
 			toolCallArgsByID[callID] += streamResp.Delta
 			if !sendToolCallDelta(callID, "", streamResp.Delta) {
-				return false
+				sr.Stop(streamErr)
+				return
 			}
 
 		case "response.function_call_arguments.done":
@@ -467,7 +476,8 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 			}
 
 			if !sendStartIfNeeded() {
-				return false
+				sr.Stop(streamErr)
+				return
 			}
 			if !sentStop {
 				if info.RelayFormat == types.RelayFormatClaude && info.ClaudeConvertInfo != nil {
@@ -479,7 +489,8 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 				}
 				stop := helper.GenerateStopResponse(responseId, createAt, model, finishReason)
 				if !sendChatChunk(stop) {
-					return false
+					sr.Stop(streamErr)
+					return
 				}
 				sentStop = true
 			}
@@ -488,16 +499,16 @@ func OaiResponsesToChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo
 			if streamResp.Response != nil {
 				if oaiErr := streamResp.Response.GetOpenAIError(); oaiErr != nil && oaiErr.Type != "" {
 					streamErr = types.WithOpenAIError(*oaiErr, http.StatusInternalServerError)
-					return false
+					sr.Stop(streamErr)
+					return
 				}
 			}
 			streamErr = types.NewOpenAIError(fmt.Errorf("responses stream error: %s", streamResp.Type), types.ErrorCodeBadResponse, http.StatusInternalServerError)
-			return false
+			sr.Stop(streamErr)
+			return
 
 		default:
 		}
-
-		return true
 	})
 
 	if streamErr != nil {

relay/channel/openai/relay-openai.go

@@ -126,11 +126,11 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 	// 检查是否为音频模型
 	isAudioModel := strings.Contains(strings.ToLower(model), "audio")
 
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		if lastStreamData != "" {
-			err := HandleStreamFormat(c, info, lastStreamData, info.ChannelSetting.ForceFormat, info.ChannelSetting.ThinkingToContent)
-			if err != nil {
+			if err := HandleStreamFormat(c, info, lastStreamData, info.ChannelSetting.ForceFormat, info.ChannelSetting.ThinkingToContent); err != nil {
 				common.SysLog("error handling stream format: " + err.Error())
+				sr.Error(err)
 			}
 		}
 		if len(data) > 0 {
@@ -142,7 +142,6 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 			lastStreamData = data
 			streamItems = append(streamItems, data)
 		}
-		return true
 	})
 
 	// 对音频模型，从倒数第二个stream data中提取usage信息
@@ -627,6 +626,12 @@ func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, res
 				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
 			}
 		}
+	case constant.ChannelTypeOpenAI:
+		if usage.PromptTokensDetails.CachedTokens == 0 {
+			if cachedTokens, ok := extractLlamaCachedTokensFromBody(responseBody); ok {
+				usage.PromptTokensDetails.CachedTokens = cachedTokens
+			}
+		}
 	}
 }
 
@@ -689,3 +694,25 @@ func extractMoonshotCachedTokensFromBody(body []byte) (int, bool) {
 
 	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_responses.go

@@ -79,55 +79,55 @@ func OaiResponsesStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp
 	var usage = &dto.Usage{}
 	var responseTextBuilder strings.Builder
 
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 
 		// 检查当前数据是否包含 completed 状态和 usage 信息
 		var streamResponse dto.ResponsesStreamResponse
-		if err := common.UnmarshalJsonStr(data, &streamResponse); err == nil {
-			sendResponsesStreamData(c, streamResponse, data)
-			switch streamResponse.Type {
-			case "response.completed":
-				if streamResponse.Response != nil {
-					if streamResponse.Response.Usage != nil {
-						if streamResponse.Response.Usage.InputTokens != 0 {
-							usage.PromptTokens = streamResponse.Response.Usage.InputTokens
-						}
-						if streamResponse.Response.Usage.OutputTokens != 0 {
-							usage.CompletionTokens = streamResponse.Response.Usage.OutputTokens
-						}
-						if streamResponse.Response.Usage.TotalTokens != 0 {
-							usage.TotalTokens = streamResponse.Response.Usage.TotalTokens
-						}
-						if streamResponse.Response.Usage.InputTokensDetails != nil {
-							usage.PromptTokensDetails.CachedTokens = streamResponse.Response.Usage.InputTokensDetails.CachedTokens
-						}
+		if err := common.UnmarshalJsonStr(data, &streamResponse); err != nil {
+			logger.LogError(c, "failed to unmarshal stream response: "+err.Error())
+			sr.Error(err)
+			return
+		}
+		sendResponsesStreamData(c, streamResponse, data)
+		switch streamResponse.Type {
+		case "response.completed":
+			if streamResponse.Response != nil {
+				if streamResponse.Response.Usage != nil {
+					if streamResponse.Response.Usage.InputTokens != 0 {
+						usage.PromptTokens = streamResponse.Response.Usage.InputTokens
 					}
-					if streamResponse.Response.HasImageGenerationCall() {
-						c.Set("image_generation_call", true)
-						c.Set("image_generation_call_quality", streamResponse.Response.GetQuality())
-						c.Set("image_generation_call_size", streamResponse.Response.GetSize())
+					if streamResponse.Response.Usage.OutputTokens != 0 {
+						usage.CompletionTokens = streamResponse.Response.Usage.OutputTokens
+					}
+					if streamResponse.Response.Usage.TotalTokens != 0 {
+						usage.TotalTokens = streamResponse.Response.Usage.TotalTokens
+					}
+					if streamResponse.Response.Usage.InputTokensDetails != nil {
+						usage.PromptTokensDetails.CachedTokens = streamResponse.Response.Usage.InputTokensDetails.CachedTokens
 					}
 				}
-			case "response.output_text.delta":
-				// 处理输出文本
-				responseTextBuilder.WriteString(streamResponse.Delta)
-			case dto.ResponsesOutputTypeItemDone:
-				// 函数调用处理
-				if streamResponse.Item != nil {
-					switch streamResponse.Item.Type {
-					case dto.BuildInCallWebSearchCall:
-						if info != nil && info.ResponsesUsageInfo != nil && info.ResponsesUsageInfo.BuiltInTools != nil {
-							if webSearchTool, exists := info.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolWebSearchPreview]; exists && webSearchTool != nil {
-								webSearchTool.CallCount++
-							}
+				if streamResponse.Response.HasImageGenerationCall() {
+					c.Set("image_generation_call", true)
+					c.Set("image_generation_call_quality", streamResponse.Response.GetQuality())
+					c.Set("image_generation_call_size", streamResponse.Response.GetSize())
+				}
+			}
+		case "response.output_text.delta":
+			// 处理输出文本
+			responseTextBuilder.WriteString(streamResponse.Delta)
+		case dto.ResponsesOutputTypeItemDone:
+			// 函数调用处理
+			if streamResponse.Item != nil {
+				switch streamResponse.Item.Type {
+				case dto.BuildInCallWebSearchCall:
+					if info != nil && info.ResponsesUsageInfo != nil && info.ResponsesUsageInfo.BuiltInTools != nil {
+						if webSearchTool, exists := info.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolWebSearchPreview]; exists && webSearchTool != nil {
+							webSearchTool.CallCount++
 						}
 					}
 				}
 			}
-		} else {
-			logger.LogError(c, "failed to unmarshal stream response: "+err.Error())
 		}
-		return true
 	})
 
 	if usage.CompletionTokens == 0 {

relay/channel/task/ali/adaptor.go

@@ -80,9 +80,9 @@ type AliVideoOutput struct {
 
 // AliUsage 使用统计
 type AliUsage struct {
-	Duration   int `json:"duration,omitempty"`
-	VideoCount int `json:"video_count,omitempty"`
-	SR         int `json:"SR,omitempty"`
+	Duration   dto.IntValue `json:"duration,omitempty"`
+	VideoCount dto.IntValue `json:"video_count,omitempty"`
+	SR         dto.IntValue `json:"SR,omitempty"`
 }
 
 type AliMetadata struct {

relay/channel/task/doubao/adaptor.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"strconv"
 	"time"
 
 	"github.com/QuantumNous/new-api/common"
@@ -13,12 +14,13 @@ import (
 	"github.com/QuantumNous/new-api/dto"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/relay/channel"
-	taskcommon "github.com/QuantumNous/new-api/relay/channel/task/taskcommon"
+	"github.com/QuantumNous/new-api/relay/channel/task/taskcommon"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	"github.com/QuantumNous/new-api/service"
 
 	"github.com/gin-gonic/gin"
 	"github.com/pkg/errors"
+	"github.com/samber/lo"
 )
 
 // ============================
@@ -26,37 +28,37 @@ import (
 // ============================
 
 type ContentItem struct {
-	Type     string          `json:"type"`                // "text", "image_url" or "video"
-	Text     string          `json:"text,omitempty"`      // for text type
-	ImageURL *ImageURL       `json:"image_url,omitempty"` // for image_url type
-	Video    *VideoReference `json:"video,omitempty"`     // for video (sample) type
-	Role     string          `json:"role,omitempty"`      // reference_image / first_frame / last_frame
+	Type     string    `json:"type,omitempty"`
+	Text     string    `json:"text,omitempty"`
+	ImageURL *MediaURL `json:"image_url,omitempty"`
+	VideoURL *MediaURL `json:"video_url,omitempty"`
+	AudioURL *MediaURL `json:"audio_url,omitempty"`
+	Role     string    `json:"role,omitempty"`
 }
 
-type ImageURL struct {
-	URL string `json:"url"`
-}
-
-type VideoReference struct {
-	URL string `json:"url"` // Draft video URL
+type MediaURL struct {
+	URL string `json:"url,omitempty"`
 }
 
 type requestPayload struct {
 	Model                 string         `json:"model"`
-	Content               []ContentItem  `json:"content"`
+	Content               []ContentItem  `json:"content,omitempty"`
 	CallbackURL           string         `json:"callback_url,omitempty"`
 	ReturnLastFrame       *dto.BoolValue `json:"return_last_frame,omitempty"`
 	ServiceTier           string         `json:"service_tier,omitempty"`
-	ExecutionExpiresAfter dto.IntValue   `json:"execution_expires_after,omitempty"`
+	ExecutionExpiresAfter *dto.IntValue  `json:"execution_expires_after,omitempty"`
 	GenerateAudio         *dto.BoolValue `json:"generate_audio,omitempty"`
 	Draft                 *dto.BoolValue `json:"draft,omitempty"`
-	Resolution            string         `json:"resolution,omitempty"`
-	Ratio                 string         `json:"ratio,omitempty"`
-	Duration              dto.IntValue   `json:"duration,omitempty"`
-	Frames                dto.IntValue   `json:"frames,omitempty"`
-	Seed                  dto.IntValue   `json:"seed,omitempty"`
-	CameraFixed           *dto.BoolValue `json:"camera_fixed,omitempty"`
-	Watermark             *dto.BoolValue `json:"watermark,omitempty"`
+	Tools                 []struct {
+		Type string `json:"type,omitempty"`
+	} `json:"tools,omitempty"`
+	Resolution  string         `json:"resolution,omitempty"`
+	Ratio       string         `json:"ratio,omitempty"`
+	Duration    *dto.IntValue  `json:"duration,omitempty"`
+	Frames      *dto.IntValue  `json:"frames,omitempty"`
+	Seed        *dto.IntValue  `json:"seed,omitempty"`
+	CameraFixed *dto.BoolValue `json:"camera_fixed,omitempty"`
+	Watermark   *dto.BoolValue `json:"watermark,omitempty"`
 }
 
 type responsePayload struct {
@@ -76,10 +78,20 @@ type responseTask struct {
 	Ratio           string `json:"ratio"`
 	FramesPerSecond int    `json:"framespersecond"`
 	ServiceTier     string `json:"service_tier"`
-	Usage           struct {
+	Tools           []struct {
+		Type string `json:"type"`
+	} `json:"tools"`
+	Usage struct {
 		CompletionTokens int `json:"completion_tokens"`
 		TotalTokens      int `json:"total_tokens"`
+		ToolUsage        struct {
+			WebSearch int `json:"web_search"`
+		} `json:"tool_usage"`
 	} `json:"usage"`
+	Error struct {
+		Code    string `json:"code"`
+		Message string `json:"message"`
+	} `json:"error"`
 	CreatedAt int64 `json:"created_at"`
 	UpdatedAt int64 `json:"updated_at"`
 }
@@ -108,18 +120,61 @@ func (a *TaskAdaptor) ValidateRequestAndSetAction(c *gin.Context, info *relaycom
 }
 
 // BuildRequestURL constructs the upstream URL.
-func (a *TaskAdaptor) BuildRequestURL(info *relaycommon.RelayInfo) (string, error) {
+func (a *TaskAdaptor) BuildRequestURL(_ *relaycommon.RelayInfo) (string, error) {
 	return fmt.Sprintf("%s/api/v3/contents/generations/tasks", a.baseURL), nil
 }
 
 // BuildRequestHeader sets required headers.
-func (a *TaskAdaptor) BuildRequestHeader(c *gin.Context, req *http.Request, info *relaycommon.RelayInfo) error {
+func (a *TaskAdaptor) BuildRequestHeader(_ *gin.Context, req *http.Request, _ *relaycommon.RelayInfo) error {
 	req.Header.Set("Content-Type", "application/json")
 	req.Header.Set("Accept", "application/json")
 	req.Header.Set("Authorization", "Bearer "+a.apiKey)
 	return nil
 }
 
+// EstimateBilling 检测请求 metadata 中是否包含视频输入，返回视频折扣 OtherRatio。
+func (a *TaskAdaptor) EstimateBilling(c *gin.Context, info *relaycommon.RelayInfo) map[string]float64 {
+	req, err := relaycommon.GetTaskRequest(c)
+	if err != nil {
+		return nil
+	}
+	if hasVideoInMetadata(req.Metadata) {
+		if ratio, ok := GetVideoInputRatio(info.OriginModelName); ok {
+			return map[string]float64{"video_input": ratio}
+		}
+	}
+	return nil
+}
+
+// hasVideoInMetadata 直接检查 metadata 的 content 数组是否包含 video_url 条目，
+// 避免构建完整的上游 requestPayload。
+func hasVideoInMetadata(metadata map[string]interface{}) bool {
+	if metadata == nil {
+		return false
+	}
+	contentRaw, ok := metadata["content"]
+	if !ok {
+		return false
+	}
+	contentSlice, ok := contentRaw.([]interface{})
+	if !ok {
+		return false
+	}
+	for _, item := range contentSlice {
+		itemMap, ok := item.(map[string]interface{})
+		if !ok {
+			continue
+		}
+		if itemMap["type"] == "video_url" {
+			return true
+		}
+		if _, has := itemMap["video_url"]; has {
+			return true
+		}
+	}
+	return false
+}
+
 // BuildRequestBody converts request into Doubao specific format.
 func (a *TaskAdaptor) BuildRequestBody(c *gin.Context, info *relaycommon.RelayInfo) (io.Reader, error) {
 	req, err := relaycommon.GetTaskRequest(c)
@@ -218,20 +273,12 @@ func (a *TaskAdaptor) convertToRequestPayload(req *relaycommon.TaskSubmitReq) (*
 		Content: []ContentItem{},
 	}
 
-	// Add text prompt
-	if req.Prompt != "" {
-		r.Content = append(r.Content, ContentItem{
-			Type: "text",
-			Text: req.Prompt,
-		})
-	}
-
 	// Add images if present
 	if req.HasImage() {
 		for _, imgURL := range req.Images {
 			r.Content = append(r.Content, ContentItem{
 				Type: "image_url",
-				ImageURL: &ImageURL{
+				ImageURL: &MediaURL{
 					URL: imgURL,
 				},
 			})
@@ -243,6 +290,16 @@ func (a *TaskAdaptor) convertToRequestPayload(req *relaycommon.TaskSubmitReq) (*
 		return nil, errors.Wrap(err, "unmarshal metadata failed")
 	}
 
+	if sec, _ := strconv.Atoi(req.Seconds); sec > 0 {
+		r.Duration = lo.ToPtr(dto.IntValue(sec))
+	}
+
+	r.Content = lo.Reject(r.Content, func(c ContentItem, _ int) bool { return c.Type == "text" })
+	r.Content = append(r.Content, ContentItem{
+		Type: "text",
+		Text: req.Prompt,
+	})
+
 	return &r, nil
 }
 
@@ -274,7 +331,7 @@ func (a *TaskAdaptor) ParseTaskResult(respBody []byte) (*relaycommon.TaskInfo, e
 	case "failed":
 		taskResult.Status = model.TaskStatusFailure
 		taskResult.Progress = "100%"
-		taskResult.Reason = "task failed"
+		taskResult.Reason = resTask.Error.Message
 	default:
 		// Unknown status, treat as processing
 		taskResult.Status = model.TaskStatusInProgress
@@ -302,8 +359,8 @@ func (a *TaskAdaptor) ConvertToOpenAIVideo(originTask *model.Task) ([]byte, erro
 
 	if dResp.Status == "failed" {
 		openAIVideo.Error = &dto.OpenAIVideoError{
-			Message: "task failed",
-			Code:    "failed",
+			Message: dResp.Error.Message,
+			Code:    dResp.Error.Code,
 		}
 	}
 

relay/channel/task/doubao/constants.go

@@ -5,6 +5,21 @@ var ModelList = []string{
 	"doubao-seedance-1-0-lite-t2v",
 	"doubao-seedance-1-0-lite-i2v",
 	"doubao-seedance-1-5-pro-251215",
+	"doubao-seedance-2-0-260128",
+	"doubao-seedance-2-0-fast-260128",
 }
 
 var ChannelName = "doubao-video"
+
+// videoInputRatioMap 视频输入折扣比率（含视频单价 / 不含视频单价）。
+// 管理员应将 ModelRatio 设置为"不含视频"的较高费率，
+// 系统在检测到视频输入时自动乘以此折扣。
+var videoInputRatioMap = map[string]float64{
+	"doubao-seedance-2-0-260128":      28.0 / 46.0, // ~0.6087
+	"doubao-seedance-2-0-fast-260128": 22.0 / 37.0, // ~0.5946
+}
+
+func GetVideoInputRatio(modelName string) (float64, bool) {
+	r, ok := videoInputRatioMap[modelName]
+	return r, ok
+}

relay/channel/task/taskcommon/helpers.go

@@ -17,6 +17,8 @@ func UnmarshalMetadata(metadata map[string]any, target any) error {
 	if metadata == nil {
 		return nil
 	}
+	// Prevent metadata from overriding model fields to avoid billing bypass.
+	delete(metadata, "model")
 	metaBytes, err := common.Marshal(metadata)
 	if err != nil {
 		return fmt.Errorf("marshal metadata failed: %w", err)

relay/channel/xai/adaptor.go

@@ -76,7 +76,7 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 	if strings.HasPrefix(request.Model, "grok-3-mini") {
 		if lo.FromPtrOr(request.MaxCompletionTokens, uint(0)) == 0 && lo.FromPtrOr(request.MaxTokens, uint(0)) != 0 {
 			request.MaxCompletionTokens = request.MaxTokens
-			request.MaxTokens = lo.ToPtr(uint(0))
+			request.MaxTokens = nil
 		}
 		if strings.HasSuffix(request.Model, "-high") {
 			request.ReasoningEffort = "high"

relay/channel/xai/text.go

@@ -43,12 +43,12 @@ func xAIStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 
 	helper.SetEventStreamHeaders(c)
 
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+	helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
 		var xAIResp *dto.ChatCompletionsStreamResponse
-		err := common.UnmarshalJsonStr(data, &xAIResp)
-		if err != nil {
+		if err := common.UnmarshalJsonStr(data, &xAIResp); err != nil {
 			common.SysLog("error unmarshalling stream response: " + err.Error())
-			return true
+			sr.Error(err)
+			return
 		}
 
 		// 把 xAI 的usage转换为 OpenAI 的usage
@@ -61,11 +61,10 @@ func xAIStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 
 		openaiResponse := streamResponseXAI2OpenAI(xAIResp, usage)
 		_ = openai.ProcessStreamResponse(*openaiResponse, &responseTextBuilder, &toolCount)
-		err = helper.ObjectData(c, openaiResponse)
-		if err != nil {
+		if err := helper.ObjectData(c, openaiResponse); err != nil {
 			common.SysLog(err.Error())
+			sr.Error(err)
 		}
-		return true
 	})
 
 	if !containStreamUsage {

relay/channel/zhipu_4v/adaptor.go

@@ -64,6 +64,9 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			}
 			return fmt.Sprintf("%s/api/paas/v4/embeddings", baseURL), nil
 		case relayconstant.RelayModeImagesGenerations:
+			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {
+				return fmt.Sprintf("%s/images/generations", specialPlan.OpenAIBaseURL), nil
+			}
 			return fmt.Sprintf("%s/api/paas/v4/images/generations", baseURL), nil
 		default:
 			if hasSpecialPlan && specialPlan.OpenAIBaseURL != "" {

relay/claude_handler.go

@@ -122,7 +122,7 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 			return newApiErr
 		}
 
-		service.PostClaudeConsumeQuota(c, info, usage)
+		service.PostTextConsumeQuota(c, info, usage, nil)
 		return nil
 	}
 
@@ -190,6 +190,6 @@ func ClaudeHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		return newAPIError
 	}
 
-	service.PostClaudeConsumeQuota(c, info, usage.(*dto.Usage))
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	return nil
 }

relay/common/relay_info.go

@@ -4,6 +4,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"strconv"
 	"strings"
 	"time"
 
@@ -162,6 +163,8 @@ type RelayInfo struct {
 	// 若为空，调用 GetFinalRequestRelayFormat 会回退到 RequestConversionChain 的最后一项或 RelayFormat。
 	FinalRequestRelayFormat types.RelayFormat
 
+	StreamStatus *StreamStatus
+
 	ThinkingContentInfo
 	TokenCountMeta
 	*ClaudeConvertInfo
@@ -338,15 +341,10 @@ func GenRelayInfoClaude(c *gin.Context, request dto.Request) *RelayInfo {
 	info.ClaudeConvertInfo = &ClaudeConvertInfo{
 		LastMessagesType: LastMessageTypeNone,
 	}
-	info.IsClaudeBetaQuery = c.Query("beta") == "true" || isClaudeBetaForced(c)
+	info.IsClaudeBetaQuery = c.Query("beta") == "true"
 	return info
 }
 
-func isClaudeBetaForced(c *gin.Context) bool {
-	channelOtherSettings, ok := common.GetContextKeyType[dto.ChannelOtherSettings](c, constant.ContextKeyChannelOtherSetting)
-	return ok && channelOtherSettings.ClaudeBetaQuery
-}
-
 func GenRelayInfoRerank(c *gin.Context, request *dto.RerankRequest) *RelayInfo {
 	info := genBaseRelayInfo(c, request)
 	info.RelayMode = relayconstant.RelayModeRerank
@@ -693,6 +691,7 @@ func (t *TaskSubmitReq) UnmarshalJSON(data []byte) error {
 	type Alias TaskSubmitReq
 	aux := &struct {
 		Metadata json.RawMessage `json:"metadata,omitempty"`
+		Duration json.RawMessage `json:"duration,omitempty"`
 		*Alias
 	}{
 		Alias: (*Alias)(t),
@@ -702,6 +701,20 @@ func (t *TaskSubmitReq) UnmarshalJSON(data []byte) error {
 		return err
 	}
 
+	if len(aux.Duration) > 0 {
+		var durationInt int
+		if err := common.Unmarshal(aux.Duration, &durationInt); err == nil {
+			t.Duration = durationInt
+		} else {
+			var durationStr string
+			if err := common.Unmarshal(aux.Duration, &durationStr); err == nil && durationStr != "" {
+				if v, err := strconv.Atoi(durationStr); err == nil {
+					t.Duration = v
+				}
+			}
+		}
+	}
+
 	if len(aux.Metadata) > 0 {
 		var metadataStr string
 		if err := common.Unmarshal(aux.Metadata, &metadataStr); err == nil && metadataStr != "" {

relay/common/relay_utils.go

@@ -204,7 +204,9 @@ func ValidateBasicTaskRequest(c *gin.Context, info *RelayInfo, action string) *d
 		if err != nil {
 			return createTaskError(err, "invalid_multipart_form", http.StatusBadRequest, true)
 		}
-	} else if err := common.UnmarshalBodyReusable(c, &req); err != nil {
+	}
+	// 为了metadata字段的兼容性，统一UnmarshalBodyReusable
+	if err := common.UnmarshalBodyReusable(c, &req); err != nil {
 		return createTaskError(err, "invalid_request", http.StatusBadRequest, true)
 	}
 

relay/common/stream_status.go

@@ -0,0 +1,112 @@
+package common
+
+import (
+	"fmt"
+	"strings"
+	"sync"
+	"time"
+)
+
+type StreamEndReason string
+
+const (
+	StreamEndReasonNone        StreamEndReason = ""
+	StreamEndReasonDone        StreamEndReason = "done"
+	StreamEndReasonTimeout     StreamEndReason = "timeout"
+	StreamEndReasonClientGone  StreamEndReason = "client_gone"
+	StreamEndReasonScannerErr  StreamEndReason = "scanner_error"
+	StreamEndReasonHandlerStop StreamEndReason = "handler_stop"
+	StreamEndReasonEOF         StreamEndReason = "eof"
+	StreamEndReasonPanic       StreamEndReason = "panic"
+	StreamEndReasonPingFail    StreamEndReason = "ping_fail"
+)
+
+const maxStreamErrorEntries = 20
+
+type StreamErrorEntry struct {
+	Message   string
+	Timestamp time.Time
+}
+
+type StreamStatus struct {
+	EndReason  StreamEndReason
+	EndError   error
+	endOnce    sync.Once
+
+	mu         sync.Mutex
+	Errors     []StreamErrorEntry
+	ErrorCount int
+}
+
+func NewStreamStatus() *StreamStatus {
+	return &StreamStatus{}
+}
+
+func (s *StreamStatus) SetEndReason(reason StreamEndReason, err error) {
+	if s == nil {
+		return
+	}
+	s.endOnce.Do(func() {
+		s.EndReason = reason
+		s.EndError = err
+	})
+}
+
+func (s *StreamStatus) RecordError(msg string) {
+	if s == nil {
+		return
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	s.ErrorCount++
+	if len(s.Errors) < maxStreamErrorEntries {
+		s.Errors = append(s.Errors, StreamErrorEntry{
+			Message:   msg,
+			Timestamp: time.Now(),
+		})
+	}
+}
+
+func (s *StreamStatus) HasErrors() bool {
+	if s == nil {
+		return false
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	return s.ErrorCount > 0
+}
+
+func (s *StreamStatus) TotalErrorCount() int {
+	if s == nil {
+		return 0
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	return s.ErrorCount
+}
+
+func (s *StreamStatus) IsNormalEnd() bool {
+	if s == nil {
+		return true
+	}
+	return s.EndReason == StreamEndReasonDone ||
+		s.EndReason == StreamEndReasonEOF ||
+		s.EndReason == StreamEndReasonHandlerStop
+}
+
+func (s *StreamStatus) Summary() string {
+	if s == nil {
+		return "StreamStatus<nil>"
+	}
+	b := &strings.Builder{}
+	fmt.Fprintf(b, "reason=%s", s.EndReason)
+	if s.EndError != nil {
+		fmt.Fprintf(b, " end_error=%q", s.EndError.Error())
+	}
+	s.mu.Lock()
+	if s.ErrorCount > 0 {
+		fmt.Fprintf(b, " soft_errors=%d", s.ErrorCount)
+	}
+	s.mu.Unlock()
+	return b.String()
+}

relay/common/stream_status_test.go

@@ -0,0 +1,182 @@
+package common
+
+import (
+	"fmt"
+	"sync"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestStreamStatus_SetEndReason_FirstWins(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	s.SetEndReason(StreamEndReasonDone, nil)
+	s.SetEndReason(StreamEndReasonTimeout, nil)
+	s.SetEndReason(StreamEndReasonClientGone, fmt.Errorf("context canceled"))
+
+	assert.Equal(t, StreamEndReasonDone, s.EndReason)
+	assert.Nil(t, s.EndError)
+}
+
+func TestStreamStatus_SetEndReason_WithError(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	expectedErr := fmt.Errorf("read: connection reset")
+	s.SetEndReason(StreamEndReasonScannerErr, expectedErr)
+
+	assert.Equal(t, StreamEndReasonScannerErr, s.EndReason)
+	assert.Equal(t, expectedErr, s.EndError)
+}
+
+func TestStreamStatus_SetEndReason_NilSafe(t *testing.T) {
+	t.Parallel()
+	var s *StreamStatus
+	s.SetEndReason(StreamEndReasonDone, nil)
+}
+
+func TestStreamStatus_SetEndReason_Concurrent(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	reasons := []StreamEndReason{
+		StreamEndReasonDone,
+		StreamEndReasonTimeout,
+		StreamEndReasonClientGone,
+		StreamEndReasonScannerErr,
+		StreamEndReasonHandlerStop,
+		StreamEndReasonEOF,
+		StreamEndReasonPanic,
+		StreamEndReasonPingFail,
+	}
+
+	var wg sync.WaitGroup
+	for _, r := range reasons {
+		wg.Add(1)
+		go func(reason StreamEndReason) {
+			defer wg.Done()
+			s.SetEndReason(reason, nil)
+		}(r)
+	}
+	wg.Wait()
+
+	assert.NotEqual(t, StreamEndReasonNone, s.EndReason)
+}
+
+func TestStreamStatus_RecordError_Basic(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	s.RecordError("bad json")
+	s.RecordError("another bad json")
+	s.RecordError("client gone")
+
+	assert.True(t, s.HasErrors())
+	assert.Equal(t, 3, s.TotalErrorCount())
+	assert.Len(t, s.Errors, 3)
+}
+
+func TestStreamStatus_RecordError_CapAtMax(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	for i := 0; i < 30; i++ {
+		s.RecordError(fmt.Sprintf("error_%d", i))
+	}
+
+	assert.Equal(t, maxStreamErrorEntries, len(s.Errors))
+	assert.Equal(t, 30, s.TotalErrorCount())
+}
+
+func TestStreamStatus_RecordError_NilSafe(t *testing.T) {
+	t.Parallel()
+	var s *StreamStatus
+	s.RecordError("should not panic")
+}
+
+func TestStreamStatus_RecordError_Concurrent(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+
+	var wg sync.WaitGroup
+	for i := 0; i < 100; i++ {
+		wg.Add(1)
+		go func(idx int) {
+			defer wg.Done()
+			s.RecordError(fmt.Sprintf("error_%d", idx))
+		}(i)
+	}
+	wg.Wait()
+
+	assert.Equal(t, 100, s.TotalErrorCount())
+	assert.LessOrEqual(t, len(s.Errors), maxStreamErrorEntries)
+}
+
+func TestStreamStatus_HasErrors_Empty(t *testing.T) {
+	t.Parallel()
+	s := NewStreamStatus()
+	assert.False(t, s.HasErrors())
+	assert.Equal(t, 0, s.TotalErrorCount())
+}
+
+func TestStreamStatus_HasErrors_NilSafe(t *testing.T) {
+	t.Parallel()
+	var s *StreamStatus
+	assert.False(t, s.HasErrors())
+	assert.Equal(t, 0, s.TotalErrorCount())
+}
+
+func TestStreamStatus_IsNormalEnd(t *testing.T) {
+	t.Parallel()
+	tests := []struct {
+		reason StreamEndReason
+		normal bool
+	}{
+		{StreamEndReasonDone, true},
+		{StreamEndReasonEOF, true},
+		{StreamEndReasonHandlerStop, true},
+		{StreamEndReasonTimeout, false},
+		{StreamEndReasonClientGone, false},
+		{StreamEndReasonScannerErr, false},
+		{StreamEndReasonPanic, false},
+		{StreamEndReasonPingFail, false},
+		{StreamEndReasonNone, false},
+	}
+	for _, tt := range tests {
+		s := NewStreamStatus()
+		s.SetEndReason(tt.reason, nil)
+		assert.Equal(t, tt.normal, s.IsNormalEnd(), "reason=%s", tt.reason)
+	}
+}
+
+func TestStreamStatus_IsNormalEnd_NilSafe(t *testing.T) {
+	t.Parallel()
+	var s *StreamStatus
+	assert.True(t, s.IsNormalEnd())
+}
+
+func TestStreamStatus_Summary(t *testing.T) {
+	t.Parallel()
+
+	s := NewStreamStatus()
+	s.SetEndReason(StreamEndReasonDone, nil)
+	summary := s.Summary()
+	assert.Contains(t, summary, "reason=done")
+	assert.NotContains(t, summary, "soft_errors")
+
+	s2 := NewStreamStatus()
+	s2.SetEndReason(StreamEndReasonTimeout, nil)
+	s2.RecordError("bad json")
+	s2.RecordError("write failed")
+	summary2 := s2.Summary()
+	assert.Contains(t, summary2, "reason=timeout")
+	assert.Contains(t, summary2, "soft_errors=2")
+}
+
+func TestStreamStatus_Summary_NilSafe(t *testing.T) {
+	t.Parallel()
+	var s *StreamStatus
+	assert.Equal(t, "StreamStatus<nil>", s.Summary())
+}

relay/compatible_handler.go

@@ -6,25 +6,20 @@ import (
 	"io"
 	"net/http"
 	"strings"
-	"time"
 
 	"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/model"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
 	relayconstant "github.com/QuantumNous/new-api/relay/constant"
 	"github.com/QuantumNous/new-api/relay/helper"
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/setting/model_setting"
-	"github.com/QuantumNous/new-api/setting/operation_setting"
 	"github.com/QuantumNous/new-api/setting/ratio_setting"
 	"github.com/QuantumNous/new-api/types"
 	"github.com/samber/lo"
 
-	"github.com/shopspring/decimal"
-
 	"github.com/gin-gonic/gin"
 )
 
@@ -93,7 +88,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 		if containAudioTokens && containsAudioRatios {
 			service.PostAudioConsumeQuota(c, info, usage, "")
 		} else {
-			postConsumeQuota(c, info, usage)
+			service.PostTextConsumeQuota(c, info, usage, nil)
 		}
 		return nil
 	}
@@ -216,293 +211,7 @@ func TextHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *types
 	if containAudioTokens && containsAudioRatios {
 		service.PostAudioConsumeQuota(c, info, usage.(*dto.Usage), "")
 	} else {
-		postConsumeQuota(c, info, usage.(*dto.Usage))
+		service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	}
 	return nil
 }
-
-func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.Usage, extraContent ...string) {
-	originUsage := usage
-	if usage == nil {
-		usage = &dto.Usage{
-			PromptTokens:     relayInfo.GetEstimatePromptTokens(),
-			CompletionTokens: 0,
-			TotalTokens:      relayInfo.GetEstimatePromptTokens(),
-		}
-		extraContent = append(extraContent, "上游无计费信息")
-	}
-
-	if originUsage != nil {
-		service.ObserveChannelAffinityUsageCacheByRelayFormat(ctx, usage, relayInfo.GetFinalRequestRelayFormat())
-	}
-
-	adminRejectReason := common.GetContextKeyString(ctx, constant.ContextKeyAdminRejectReason)
-
-	useTimeSeconds := time.Now().Unix() - relayInfo.StartTime.Unix()
-	promptTokens := usage.PromptTokens
-	cacheTokens := usage.PromptTokensDetails.CachedTokens
-	imageTokens := usage.PromptTokensDetails.ImageTokens
-	audioTokens := usage.PromptTokensDetails.AudioTokens
-	completionTokens := usage.CompletionTokens
-	cachedCreationTokens := usage.PromptTokensDetails.CachedCreationTokens
-
-	modelName := relayInfo.OriginModelName
-
-	tokenName := ctx.GetString("token_name")
-	completionRatio := relayInfo.PriceData.CompletionRatio
-	cacheRatio := relayInfo.PriceData.CacheRatio
-	imageRatio := relayInfo.PriceData.ImageRatio
-	modelRatio := relayInfo.PriceData.ModelRatio
-	groupRatio := relayInfo.PriceData.GroupRatioInfo.GroupRatio
-	modelPrice := relayInfo.PriceData.ModelPrice
-	cachedCreationRatio := relayInfo.PriceData.CacheCreationRatio
-
-	// Convert values to decimal for precise calculation
-	dPromptTokens := decimal.NewFromInt(int64(promptTokens))
-	dCacheTokens := decimal.NewFromInt(int64(cacheTokens))
-	dImageTokens := decimal.NewFromInt(int64(imageTokens))
-	dAudioTokens := decimal.NewFromInt(int64(audioTokens))
-	dCompletionTokens := decimal.NewFromInt(int64(completionTokens))
-	dCachedCreationTokens := decimal.NewFromInt(int64(cachedCreationTokens))
-	dCompletionRatio := decimal.NewFromFloat(completionRatio)
-	dCacheRatio := decimal.NewFromFloat(cacheRatio)
-	dImageRatio := decimal.NewFromFloat(imageRatio)
-	dModelRatio := decimal.NewFromFloat(modelRatio)
-	dGroupRatio := decimal.NewFromFloat(groupRatio)
-	dModelPrice := decimal.NewFromFloat(modelPrice)
-	dCachedCreationRatio := decimal.NewFromFloat(cachedCreationRatio)
-	dQuotaPerUnit := decimal.NewFromFloat(common.QuotaPerUnit)
-
-	ratio := dModelRatio.Mul(dGroupRatio)
-
-	// openai web search 工具计费
-	var dWebSearchQuota decimal.Decimal
-	var webSearchPrice float64
-	// response api 格式工具计费
-	if relayInfo.ResponsesUsageInfo != nil {
-		if webSearchTool, exists := relayInfo.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolWebSearchPreview]; exists && webSearchTool.CallCount > 0 {
-			// 计算 web search 调用的配额 (配额 = 价格 * 调用次数 / 1000 * 分组倍率)
-			webSearchPrice = operation_setting.GetWebSearchPricePerThousand(modelName, webSearchTool.SearchContextSize)
-			dWebSearchQuota = decimal.NewFromFloat(webSearchPrice).
-				Mul(decimal.NewFromInt(int64(webSearchTool.CallCount))).
-				Div(decimal.NewFromInt(1000)).Mul(dGroupRatio).Mul(dQuotaPerUnit)
-			extraContent = append(extraContent, fmt.Sprintf("Web Search 调用 %d 次，上下文大小 %s，调用花费 %s",
-				webSearchTool.CallCount, webSearchTool.SearchContextSize, dWebSearchQuota.String()))
-		}
-	} else if strings.HasSuffix(modelName, "search-preview") {
-		// search-preview 模型不支持 response api
-		searchContextSize := ctx.GetString("chat_completion_web_search_context_size")
-		if searchContextSize == "" {
-			searchContextSize = "medium"
-		}
-		webSearchPrice = operation_setting.GetWebSearchPricePerThousand(modelName, searchContextSize)
-		dWebSearchQuota = decimal.NewFromFloat(webSearchPrice).
-			Div(decimal.NewFromInt(1000)).Mul(dGroupRatio).Mul(dQuotaPerUnit)
-		extraContent = append(extraContent, fmt.Sprintf("Web Search 调用 1 次，上下文大小 %s，调用花费 %s",
-			searchContextSize, dWebSearchQuota.String()))
-	}
-	// claude web search tool 计费
-	var dClaudeWebSearchQuota decimal.Decimal
-	var claudeWebSearchPrice float64
-	claudeWebSearchCallCount := ctx.GetInt("claude_web_search_requests")
-	if claudeWebSearchCallCount > 0 {
-		claudeWebSearchPrice = operation_setting.GetClaudeWebSearchPricePerThousand()
-		dClaudeWebSearchQuota = decimal.NewFromFloat(claudeWebSearchPrice).
-			Div(decimal.NewFromInt(1000)).Mul(dGroupRatio).Mul(dQuotaPerUnit).Mul(decimal.NewFromInt(int64(claudeWebSearchCallCount)))
-		extraContent = append(extraContent, fmt.Sprintf("Claude Web Search 调用 %d 次，调用花费 %s",
-			claudeWebSearchCallCount, dClaudeWebSearchQuota.String()))
-	}
-	// file search tool 计费
-	var dFileSearchQuota decimal.Decimal
-	var fileSearchPrice float64
-	if relayInfo.ResponsesUsageInfo != nil {
-		if fileSearchTool, exists := relayInfo.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolFileSearch]; exists && fileSearchTool.CallCount > 0 {
-			fileSearchPrice = operation_setting.GetFileSearchPricePerThousand()
-			dFileSearchQuota = decimal.NewFromFloat(fileSearchPrice).
-				Mul(decimal.NewFromInt(int64(fileSearchTool.CallCount))).
-				Div(decimal.NewFromInt(1000)).Mul(dGroupRatio).Mul(dQuotaPerUnit)
-			extraContent = append(extraContent, fmt.Sprintf("File Search 调用 %d 次，调用花费 %s",
-				fileSearchTool.CallCount, dFileSearchQuota.String()))
-		}
-	}
-	var dImageGenerationCallQuota decimal.Decimal
-	var imageGenerationCallPrice float64
-	if ctx.GetBool("image_generation_call") {
-		imageGenerationCallPrice = operation_setting.GetGPTImage1PriceOnceCall(ctx.GetString("image_generation_call_quality"), ctx.GetString("image_generation_call_size"))
-		dImageGenerationCallQuota = decimal.NewFromFloat(imageGenerationCallPrice).Mul(dGroupRatio).Mul(dQuotaPerUnit)
-		extraContent = append(extraContent, fmt.Sprintf("Image Generation Call 花费 %s", dImageGenerationCallQuota.String()))
-	}
-
-	var quotaCalculateDecimal decimal.Decimal
-
-	var audioInputQuota decimal.Decimal
-	var audioInputPrice float64
-	isClaudeUsageSemantic := relayInfo.GetFinalRequestRelayFormat() == types.RelayFormatClaude
-	if !relayInfo.PriceData.UsePrice {
-		baseTokens := dPromptTokens
-		// 减去 cached tokens
-		// Anthropic API 的 input_tokens 已经不包含缓存 tokens，不需要减去
-		// OpenAI/OpenRouter 等 API 的 prompt_tokens 包含缓存 tokens，需要减去
-		var cachedTokensWithRatio decimal.Decimal
-		if !dCacheTokens.IsZero() {
-			if !isClaudeUsageSemantic {
-				baseTokens = baseTokens.Sub(dCacheTokens)
-			}
-			cachedTokensWithRatio = dCacheTokens.Mul(dCacheRatio)
-		}
-		var dCachedCreationTokensWithRatio decimal.Decimal
-		if !dCachedCreationTokens.IsZero() {
-			if !isClaudeUsageSemantic {
-				baseTokens = baseTokens.Sub(dCachedCreationTokens)
-			}
-			dCachedCreationTokensWithRatio = dCachedCreationTokens.Mul(dCachedCreationRatio)
-		}
-
-		// 减去 image tokens
-		var imageTokensWithRatio decimal.Decimal
-		if !dImageTokens.IsZero() {
-			baseTokens = baseTokens.Sub(dImageTokens)
-			imageTokensWithRatio = dImageTokens.Mul(dImageRatio)
-		}
-
-		// 减去 Gemini audio tokens
-		if !dAudioTokens.IsZero() {
-			audioInputPrice = operation_setting.GetGeminiInputAudioPricePerMillionTokens(modelName)
-			if audioInputPrice > 0 {
-				// 重新计算 base tokens
-				baseTokens = baseTokens.Sub(dAudioTokens)
-				audioInputQuota = decimal.NewFromFloat(audioInputPrice).Div(decimal.NewFromInt(1000000)).Mul(dAudioTokens).Mul(dGroupRatio).Mul(dQuotaPerUnit)
-				extraContent = append(extraContent, fmt.Sprintf("Audio Input 花费 %s", audioInputQuota.String()))
-			}
-		}
-		promptQuota := baseTokens.Add(cachedTokensWithRatio).
-			Add(imageTokensWithRatio).
-			Add(dCachedCreationTokensWithRatio)
-
-		completionQuota := dCompletionTokens.Mul(dCompletionRatio)
-
-		quotaCalculateDecimal = promptQuota.Add(completionQuota).Mul(ratio)
-
-		if !ratio.IsZero() && quotaCalculateDecimal.LessThanOrEqual(decimal.Zero) {
-			quotaCalculateDecimal = decimal.NewFromInt(1)
-		}
-	} else {
-		quotaCalculateDecimal = dModelPrice.Mul(dQuotaPerUnit).Mul(dGroupRatio)
-	}
-	// 添加 responses tools call 调用的配额
-	quotaCalculateDecimal = quotaCalculateDecimal.Add(dWebSearchQuota)
-	quotaCalculateDecimal = quotaCalculateDecimal.Add(dFileSearchQuota)
-	// 添加 audio input 独立计费
-	quotaCalculateDecimal = quotaCalculateDecimal.Add(audioInputQuota)
-	// 添加 image generation call 计费
-	quotaCalculateDecimal = quotaCalculateDecimal.Add(dImageGenerationCallQuota)
-
-	if len(relayInfo.PriceData.OtherRatios) > 0 {
-		for key, otherRatio := range relayInfo.PriceData.OtherRatios {
-			dOtherRatio := decimal.NewFromFloat(otherRatio)
-			quotaCalculateDecimal = quotaCalculateDecimal.Mul(dOtherRatio)
-			extraContent = append(extraContent, fmt.Sprintf("其他倍率 %s: %f", key, otherRatio))
-		}
-	}
-
-	quota := int(quotaCalculateDecimal.Round(0).IntPart())
-	totalTokens := promptTokens + completionTokens
-
-	//var logContent string
-
-	// record all the consume log even if quota is 0
-	if totalTokens == 0 {
-		// in this case, must be some error happened
-		// we cannot just return, because we may have to return the pre-consumed quota
-		quota = 0
-		extraContent = append(extraContent, "上游没有返回计费信息，无法扣费（可能是上游超时）")
-		logger.LogError(ctx, fmt.Sprintf("total tokens is 0, cannot consume quota, userId %d, channelId %d, "+
-			"tokenId %d, model %s， pre-consumed quota %d", relayInfo.UserId, relayInfo.ChannelId, relayInfo.TokenId, modelName, relayInfo.FinalPreConsumedQuota))
-	} else {
-		if !ratio.IsZero() && quota == 0 {
-			quota = 1
-		}
-		model.UpdateUserUsedQuotaAndRequestCount(relayInfo.UserId, quota)
-		model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
-	}
-
-	if err := service.SettleBilling(ctx, relayInfo, quota); err != nil {
-		logger.LogError(ctx, "error settling billing: "+err.Error())
-	}
-
-	logModel := modelName
-	if strings.HasPrefix(logModel, "gpt-4-gizmo") {
-		logModel = "gpt-4-gizmo-*"
-		extraContent = append(extraContent, fmt.Sprintf("模型 %s", modelName))
-	}
-	if strings.HasPrefix(logModel, "gpt-4o-gizmo") {
-		logModel = "gpt-4o-gizmo-*"
-		extraContent = append(extraContent, fmt.Sprintf("模型 %s", modelName))
-	}
-	logContent := strings.Join(extraContent, ", ")
-	other := service.GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, cacheTokens, cacheRatio, modelPrice, relayInfo.PriceData.GroupRatioInfo.GroupSpecialRatio)
-	if adminRejectReason != "" {
-		other["reject_reason"] = adminRejectReason
-	}
-	// For chat-based calls to the Claude model, tagging is required. Using Claude's rendering logs, the two approaches handle input rendering differently.
-	if isClaudeUsageSemantic {
-		other["claude"] = true
-		other["usage_semantic"] = "anthropic"
-	}
-	if imageTokens != 0 {
-		other["image"] = true
-		other["image_ratio"] = imageRatio
-		other["image_output"] = imageTokens
-	}
-	if cachedCreationTokens != 0 {
-		other["cache_creation_tokens"] = cachedCreationTokens
-		other["cache_creation_ratio"] = cachedCreationRatio
-	}
-	if !dWebSearchQuota.IsZero() {
-		if relayInfo.ResponsesUsageInfo != nil {
-			if webSearchTool, exists := relayInfo.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolWebSearchPreview]; exists {
-				other["web_search"] = true
-				other["web_search_call_count"] = webSearchTool.CallCount
-				other["web_search_price"] = webSearchPrice
-			}
-		} else if strings.HasSuffix(modelName, "search-preview") {
-			other["web_search"] = true
-			other["web_search_call_count"] = 1
-			other["web_search_price"] = webSearchPrice
-		}
-	} else if !dClaudeWebSearchQuota.IsZero() {
-		other["web_search"] = true
-		other["web_search_call_count"] = claudeWebSearchCallCount
-		other["web_search_price"] = claudeWebSearchPrice
-	}
-	if !dFileSearchQuota.IsZero() && relayInfo.ResponsesUsageInfo != nil {
-		if fileSearchTool, exists := relayInfo.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolFileSearch]; exists {
-			other["file_search"] = true
-			other["file_search_call_count"] = fileSearchTool.CallCount
-			other["file_search_price"] = fileSearchPrice
-		}
-	}
-	if !audioInputQuota.IsZero() {
-		other["audio_input_seperate_price"] = true
-		other["audio_input_token_count"] = audioTokens
-		other["audio_input_price"] = audioInputPrice
-	}
-	if !dImageGenerationCallQuota.IsZero() {
-		other["image_generation_call"] = true
-		other["image_generation_call_price"] = imageGenerationCallPrice
-	}
-	model.RecordConsumeLog(ctx, relayInfo.UserId, model.RecordConsumeLogParams{
-		ChannelId:        relayInfo.ChannelId,
-		PromptTokens:     promptTokens,
-		CompletionTokens: completionTokens,
-		ModelName:        logModel,
-		TokenName:        tokenName,
-		Quota:            quota,
-		Content:          logContent,
-		TokenId:          relayInfo.TokenId,
-		UseTimeSeconds:   int(useTimeSeconds),
-		IsStream:         relayInfo.IsStream,
-		Group:            relayInfo.UsingGroup,
-		Other:            other,
-	})
-}

relay/embedding_handler.go

@@ -82,6 +82,6 @@ func EmbeddingHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 		service.ResetStatusCode(newAPIError, statusCodeMappingStr)
 		return newAPIError
 	}
-	postConsumeQuota(c, info, usage.(*dto.Usage))
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	return nil
 }

relay/gemini_handler.go

@@ -194,7 +194,7 @@ func GeminiHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		return openaiErr
 	}
 
-	postConsumeQuota(c, info, usage.(*dto.Usage))
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	return nil
 }
 
@@ -288,6 +288,6 @@ func GeminiEmbeddingHandler(c *gin.Context, info *relaycommon.RelayInfo) (newAPI
 		return openaiErr
 	}
 
-	postConsumeQuota(c, info, usage.(*dto.Usage))
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	return nil
 }

relay/helper/price.go

@@ -139,21 +139,23 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 	return priceData, nil
 }
 
-// ModelPriceHelperPerCall 按次计费的 PriceHelper (MJ、Task)
+// ModelPriceHelperPerCall 按次/按量计费的 PriceHelper (MJ、Task)
 func ModelPriceHelperPerCall(c *gin.Context, info *relaycommon.RelayInfo) (types.PriceData, error) {
 	groupRatioInfo := HandleGroupRatio(c, info)
 
 	modelPrice, success := ratio_setting.GetModelPrice(info.OriginModelName, true)
-	// 如果没有配置价格，检查模型倍率配置
-	if !success {
+	usePrice := success
+	var modelRatio float64
 
-		// 没有配置费用，也要使用默认费用,否则按费率计费模型无法使用
+	if !success {
 		defaultPrice, ok := ratio_setting.GetDefaultModelPriceMap()[info.OriginModelName]
 		if ok {
 			modelPrice = defaultPrice
+			usePrice = true
 		} else {
-			// 没有配置倍率也不接受没配置,那就返回错误
-			_, ratioSuccess, matchName := ratio_setting.GetModelRatio(info.OriginModelName)
+			var ratioSuccess bool
+			var matchName string
+			modelRatio, ratioSuccess, matchName = ratio_setting.GetModelRatio(info.OriginModelName)
 			acceptUnsetRatio := false
 			if info.UserSetting.AcceptUnsetRatioModel {
 				acceptUnsetRatio = true
@@ -161,25 +163,37 @@ func ModelPriceHelperPerCall(c *gin.Context, info *relaycommon.RelayInfo) (types
 			if !ratioSuccess && !acceptUnsetRatio {
 				return types.PriceData{}, fmt.Errorf("模型 %s 倍率或价格未配置，请联系管理员设置或开始自用模式；Model %s ratio or price not set, please set or start self-use mode", matchName, matchName)
 			}
-			// 未配置价格但配置了倍率，使用默认预扣价格
-			modelPrice = float64(common.PreConsumedQuota) / common.QuotaPerUnit
 		}
-
 	}
-	quota := int(modelPrice * common.QuotaPerUnit * groupRatioInfo.GroupRatio)
 
-	// 免费模型检测（与 ModelPriceHelper 对齐）
+	var quota int
 	freeModel := false
-	if !operation_setting.GetQuotaSetting().EnableFreeModelPreConsume {
-		if groupRatioInfo.GroupRatio == 0 || modelPrice == 0 {
-			quota = 0
-			freeModel = true
+
+	if usePrice {
+		quota = int(modelPrice * common.QuotaPerUnit * groupRatioInfo.GroupRatio)
+		if !operation_setting.GetQuotaSetting().EnableFreeModelPreConsume {
+			if groupRatioInfo.GroupRatio == 0 || modelPrice == 0 {
+				quota = 0
+				freeModel = true
+			}
+		}
+	} else {
+		// 按量计费：以模型倍率的一半作为预扣额度
+		quota = int(modelRatio / 2 * common.QuotaPerUnit * groupRatioInfo.GroupRatio)
+		modelPrice = -1
+		if !operation_setting.GetQuotaSetting().EnableFreeModelPreConsume {
+			if groupRatioInfo.GroupRatio == 0 || modelRatio == 0 {
+				quota = 0
+				freeModel = true
+			}
 		}
 	}
 
 	priceData := types.PriceData{
 		FreeModel:      freeModel,
 		ModelPrice:     modelPrice,
+		ModelRatio:     modelRatio,
+		UsePrice:       usePrice,
 		Quota:          quota,
 		GroupRatioInfo: groupRatioInfo,
 	}

relay/helper/stream_result.go

@@ -0,0 +1,52 @@
+package helper
+
+import (
+	relaycommon "github.com/QuantumNous/new-api/relay/common"
+)
+
+// StreamResult is passed to each dataHandler invocation, providing methods
+// to record soft errors, signal fatal stops, or mark normal completion.
+// StreamScannerHandler checks IsStopped() after each callback invocation.
+type StreamResult struct {
+	status  *relaycommon.StreamStatus
+	stopped bool
+}
+
+func newStreamResult(status *relaycommon.StreamStatus) *StreamResult {
+	return &StreamResult{status: status}
+}
+
+// Error records a soft error. The stream continues processing.
+// Can be called multiple times per chunk.
+func (r *StreamResult) Error(err error) {
+	if err == nil {
+		return
+	}
+	r.status.RecordError(err.Error())
+}
+
+// Stop records a fatal error and marks the stream to stop after this chunk.
+func (r *StreamResult) Stop(err error) {
+	if err != nil {
+		r.status.RecordError(err.Error())
+	}
+	r.status.SetEndReason(relaycommon.StreamEndReasonHandlerStop, err)
+	r.stopped = true
+}
+
+// Done signals that the handler has finished processing normally
+// (e.g., Dify "message_end"). The stream stops after this chunk.
+func (r *StreamResult) Done() {
+	r.status.SetEndReason(relaycommon.StreamEndReasonDone, nil)
+	r.stopped = true
+}
+
+// IsStopped returns whether Stop() or Done() was called during this chunk.
+func (r *StreamResult) IsStopped() bool {
+	return r.stopped
+}
+
+// reset clears the per-chunk stopped flag so the object can be reused.
+func (r *StreamResult) reset() {
+	r.stopped = false
+}

relay/helper/stream_scanner.go

@@ -34,12 +34,15 @@ func getScannerBufferSize() int {
 	return DefaultMaxScannerBufferSize
 }
 
-func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, dataHandler func(data string) bool) {
+func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, dataHandler func(data string, sr *StreamResult)) {
 
 	if resp == nil || dataHandler == nil {
 		return
 	}
 
+	// 无条件新建 StreamStatus
+	info.StreamStatus = relaycommon.NewStreamStatus()
+
 	// 确保响应体总是被关闭
 	defer func() {
 		if resp.Body != nil {
@@ -121,6 +124,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 				wg.Done()
 				if r := recover(); r != nil {
 					logger.LogError(c, fmt.Sprintf("ping goroutine panic: %v", r))
+					info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonPanic, fmt.Errorf("ping panic: %v", r))
 					common.SafeSendBool(stopChan, true)
 				}
 				if common.DebugEnabled {
@@ -148,6 +152,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 					case err := <-done:
 						if err != nil {
 							logger.LogError(c, "ping data error: "+err.Error())
+							info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonPingFail, err)
 							return
 						}
 						if common.DebugEnabled {
@@ -155,6 +160,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 						}
 					case <-time.After(10 * time.Second):
 						logger.LogError(c, "ping data send timeout")
+						info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonPingFail, fmt.Errorf("ping send timeout"))
 						return
 					case <-ctx.Done():
 						return
@@ -184,14 +190,17 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 			wg.Done()
 			if r := recover(); r != nil {
 				logger.LogError(c, fmt.Sprintf("data handler goroutine panic: %v", r))
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonPanic, fmt.Errorf("handler panic: %v", r))
 			}
 			common.SafeSendBool(stopChan, true)
 		}()
+		sr := newStreamResult(info.StreamStatus)
 		for data := range dataChan {
+			sr.reset()
 			writeMutex.Lock()
-			success := dataHandler(data)
+			dataHandler(data, sr)
 			writeMutex.Unlock()
-			if !success {
+			if sr.IsStopped() {
 				return
 			}
 		}
@@ -205,6 +214,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 			wg.Done()
 			if r := recover(); r != nil {
 				logger.LogError(c, fmt.Sprintf("scanner goroutine panic: %v", r))
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonPanic, fmt.Errorf("scanner panic: %v", r))
 			}
 			common.SafeSendBool(stopChan, true)
 			if common.DebugEnabled {
@@ -220,6 +230,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 			case <-ctx.Done():
 				return
 			case <-c.Request.Context().Done():
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, c.Request.Context().Err())
 				return
 			default:
 			}
@@ -253,7 +264,7 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 					return
 				}
 			} else {
-				// done, 处理完成标志，直接退出停止读取剩余数据防止出错
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonDone, nil)
 				if common.DebugEnabled {
 					println("received [DONE], stopping scanner")
 				}
@@ -264,20 +275,25 @@ func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon
 		if err := scanner.Err(); err != nil {
 			if err != io.EOF {
 				logger.LogError(c, "scanner error: "+err.Error())
+				info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonScannerErr, err)
 			}
 		}
+		info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonEOF, nil)
 	})
 
 	// 主循环等待完成或超时
 	select {
 	case <-ticker.C:
-		// 超时处理逻辑
-		logger.LogError(c, "streaming timeout")
+		info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonTimeout, nil)
 	case <-stopChan:
-		// 正常结束
-		logger.LogInfo(c, "streaming finished")
+		// EndReason already set by the goroutine that triggered stopChan
 	case <-c.Request.Context().Done():
-		// 客户端断开连接
-		logger.LogInfo(c, "client disconnected")
+		info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, c.Request.Context().Err())
+	}
+
+	if info.StreamStatus.IsNormalEnd() && !info.StreamStatus.HasErrors() {
+		logger.LogInfo(c, fmt.Sprintf("stream ended: %s", info.StreamStatus.Summary()))
+	} else {
+		logger.LogError(c, fmt.Sprintf("stream ended: %s, received=%d", info.StreamStatus.Summary(), info.ReceivedResponseCount))
 	}
 }

relay/helper/stream_scanner_test.go

@@ -56,8 +56,6 @@ func buildSSEBody(n int) string {
 	return b.String()
 }
 
-// slowReader wraps a reader and injects a delay before each Read call,
-// simulating a slow upstream that trickles data.
 type slowReader struct {
 	r     io.Reader
 	delay time.Duration
@@ -79,7 +77,7 @@ func TestStreamScannerHandler_NilInputs(t *testing.T) {
 
 	info := &relaycommon.RelayInfo{ChannelMeta: &relaycommon.ChannelMeta{}}
 
-	StreamScannerHandler(c, nil, info, func(data string) bool { return true })
+	StreamScannerHandler(c, nil, info, func(data string, sr *StreamResult) {})
 	StreamScannerHandler(c, &http.Response{Body: io.NopCloser(strings.NewReader(""))}, info, nil)
 }
 
@@ -89,9 +87,8 @@ func TestStreamScannerHandler_EmptyBody(t *testing.T) {
 	c, resp, info := setupStreamTest(t, strings.NewReader(""))
 
 	var called atomic.Bool
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		called.Store(true)
-		return true
 	})
 
 	assert.False(t, called.Load(), "handler should not be called for empty body")
@@ -105,9 +102,8 @@ func TestStreamScannerHandler_1000Chunks(t *testing.T) {
 	c, resp, info := setupStreamTest(t, strings.NewReader(body))
 
 	var count atomic.Int64
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		count.Add(1)
-		return true
 	})
 
 	assert.Equal(t, int64(numChunks), count.Load())
@@ -124,9 +120,8 @@ func TestStreamScannerHandler_10000Chunks(t *testing.T) {
 	var count atomic.Int64
 	start := time.Now()
 
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		count.Add(1)
-		return true
 	})
 
 	elapsed := time.Since(start)
@@ -145,11 +140,10 @@ func TestStreamScannerHandler_OrderPreserved(t *testing.T) {
 	var mu sync.Mutex
 	received := make([]string, 0, numChunks)
 
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		mu.Lock()
 		received = append(received, data)
 		mu.Unlock()
-		return true
 	})
 
 	require.Equal(t, numChunks, len(received))
@@ -166,31 +160,32 @@ func TestStreamScannerHandler_DoneStopsScanner(t *testing.T) {
 	c, resp, info := setupStreamTest(t, strings.NewReader(body))
 
 	var count atomic.Int64
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		count.Add(1)
-		return true
 	})
 
 	assert.Equal(t, int64(50), count.Load(), "data after [DONE] must not be processed")
 }
 
-func TestStreamScannerHandler_HandlerFailureStops(t *testing.T) {
+func TestStreamScannerHandler_StopStopsStream(t *testing.T) {
 	t.Parallel()
 
 	const numChunks = 200
 	body := buildSSEBody(numChunks)
 	c, resp, info := setupStreamTest(t, strings.NewReader(body))
 
-	const failAt = 50
+	const stopAt int64 = 50
 	var count atomic.Int64
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		n := count.Add(1)
-		return n < failAt
+		if n >= stopAt {
+			sr.Stop(fmt.Errorf("fatal at %d", n))
+		}
 	})
 
-	// The worker stops at failAt; the scanner may have read ahead,
-	// but the handler should not be called beyond failAt.
-	assert.Equal(t, int64(failAt), count.Load())
+	assert.Equal(t, stopAt, count.Load())
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonHandlerStop, info.StreamStatus.EndReason)
 }
 
 func TestStreamScannerHandler_SkipsNonDataLines(t *testing.T) {
@@ -210,9 +205,8 @@ func TestStreamScannerHandler_SkipsNonDataLines(t *testing.T) {
 	c, resp, info := setupStreamTest(t, strings.NewReader(b.String()))
 
 	var count atomic.Int64
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		count.Add(1)
-		return true
 	})
 
 	assert.Equal(t, int64(100), count.Load())
@@ -225,25 +219,18 @@ func TestStreamScannerHandler_DataWithExtraSpaces(t *testing.T) {
 	c, resp, info := setupStreamTest(t, strings.NewReader(body))
 
 	var got string
-	StreamScannerHandler(c, resp, info, func(data string) bool {
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 		got = data
-		return true
 	})
 
 	assert.Equal(t, "{\"trimmed\":true}", got)
 }
 
-// ---------- Decoupling: scanner not blocked by slow handler ----------
+// ---------- Decoupling ----------
 
 func TestStreamScannerHandler_ScannerDecoupledFromSlowHandler(t *testing.T) {
 	t.Parallel()
 
-	// Strategy: use a slow upstream (io.Pipe, 10ms per chunk) AND a slow handler (20ms per chunk).
-	// If the scanner were synchronously coupled to the handler, total time would be
-	// ~numChunks * (10ms + 20ms) = 30ms * 50 = 1500ms.
-	// With decoupling, total time should be closer to
-	// ~numChunks * max(10ms, 20ms) = 20ms * 50 = 1000ms
-	// because the scanner reads ahead into the buffer while the handler processes.
 	const numChunks = 50
 	const upstreamDelay = 10 * time.Millisecond
 	const handlerDelay = 20 * time.Millisecond
@@ -273,10 +260,9 @@ func TestStreamScannerHandler_ScannerDecoupledFromSlowHandler(t *testing.T) {
 	start := time.Now()
 	done := make(chan struct{})
 	go func() {
-		StreamScannerHandler(c, resp, info, func(data string) bool {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 			time.Sleep(handlerDelay)
 			count.Add(1)
-			return true
 		})
 		close(done)
 	}()
@@ -293,7 +279,6 @@ func TestStreamScannerHandler_ScannerDecoupledFromSlowHandler(t *testing.T) {
 	coupledTime := time.Duration(numChunks) * (upstreamDelay + handlerDelay)
 	t.Logf("elapsed=%v, coupled_estimate=%v", elapsed, coupledTime)
 
-	// If decoupled, elapsed should be well under the coupled estimate.
 	assert.Less(t, elapsed, coupledTime*85/100,
 		"decoupled elapsed time (%v) should be significantly less than coupled estimate (%v)", elapsed, coupledTime)
 }
@@ -311,9 +296,8 @@ func TestStreamScannerHandler_SlowUpstreamFastHandler(t *testing.T) {
 
 	done := make(chan struct{})
 	go func() {
-		StreamScannerHandler(c, resp, info, func(data string) bool {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 			count.Add(1)
-			return true
 		})
 		close(done)
 	}()
@@ -344,8 +328,6 @@ func TestStreamScannerHandler_PingSentDuringSlowUpstream(t *testing.T) {
 		setting.PingIntervalSeconds = oldSeconds
 	})
 
-	// Create a reader that delivers data slowly: one chunk every 500ms over 3.5 seconds.
-	// The ping interval is 1s, so we should see at least 2 pings.
 	pr, pw := io.Pipe()
 	go func() {
 		defer pw.Close()
@@ -372,9 +354,8 @@ func TestStreamScannerHandler_PingSentDuringSlowUpstream(t *testing.T) {
 	var count atomic.Int64
 	done := make(chan struct{})
 	go func() {
-		StreamScannerHandler(c, resp, info, func(data string) bool {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 			count.Add(1)
-			return true
 		})
 		close(done)
 	}()
@@ -436,9 +417,8 @@ func TestStreamScannerHandler_PingDisabledByRelayInfo(t *testing.T) {
 	var count atomic.Int64
 	done := make(chan struct{})
 	go func() {
-		StreamScannerHandler(c, resp, info, func(data string) bool {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 			count.Add(1)
-			return true
 		})
 		close(done)
 	}()
@@ -456,6 +436,199 @@ func TestStreamScannerHandler_PingDisabledByRelayInfo(t *testing.T) {
 	assert.Equal(t, 0, pingCount, "pings should be disabled when DisablePing=true")
 }
 
+// ---------- StreamStatus integration ----------
+
+func TestStreamScannerHandler_StreamStatus_DoneReason(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(10)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {})
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonDone, info.StreamStatus.EndReason)
+	assert.Nil(t, info.StreamStatus.EndError)
+	assert.True(t, info.StreamStatus.IsNormalEnd())
+	assert.False(t, info.StreamStatus.HasErrors())
+}
+
+func TestStreamScannerHandler_StreamStatus_EOFWithoutDone(t *testing.T) {
+	t.Parallel()
+
+	var b strings.Builder
+	for i := 0; i < 5; i++ {
+		fmt.Fprintf(&b, "data: {\"id\":%d}\n", i)
+	}
+	c, resp, info := setupStreamTest(t, strings.NewReader(b.String()))
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {})
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonEOF, info.StreamStatus.EndReason)
+	assert.True(t, info.StreamStatus.IsNormalEnd())
+}
+
+func TestStreamScannerHandler_StreamStatus_HandlerStop(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(100)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	var count atomic.Int64
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
+		n := count.Add(1)
+		if n >= 10 {
+			sr.Stop(fmt.Errorf("stop at 10"))
+		}
+	})
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonHandlerStop, info.StreamStatus.EndReason)
+	assert.True(t, info.StreamStatus.HasErrors())
+}
+
+func TestStreamScannerHandler_StreamStatus_HandlerDone(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(20)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	var count atomic.Int64
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
+		n := count.Add(1)
+		if n >= 5 {
+			sr.Done()
+		}
+	})
+
+	assert.Equal(t, int64(5), count.Load())
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonDone, info.StreamStatus.EndReason)
+	assert.False(t, info.StreamStatus.HasErrors())
+}
+
+func TestStreamScannerHandler_StreamStatus_Timeout(t *testing.T) {
+	// Not parallel: modifies global constant.StreamingTimeout
+	oldTimeout := constant.StreamingTimeout
+	constant.StreamingTimeout = 2
+	t.Cleanup(func() { constant.StreamingTimeout = oldTimeout })
+
+	pr, pw := io.Pipe()
+	go func() {
+		fmt.Fprint(pw, "data: {\"id\":1}\n")
+		time.Sleep(10 * time.Second)
+		pw.Close()
+	}()
+
+	recorder := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(recorder)
+	c.Request = httptest.NewRequest(http.MethodPost, "/v1/chat/completions", nil)
+
+	resp := &http.Response{Body: pr}
+	info := &relaycommon.RelayInfo{ChannelMeta: &relaycommon.ChannelMeta{}}
+
+	done := make(chan struct{})
+	go func() {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {})
+		close(done)
+	}()
+
+	select {
+	case <-done:
+	case <-time.After(15 * time.Second):
+		t.Fatal("timed out waiting for stream timeout")
+	}
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonTimeout, info.StreamStatus.EndReason)
+	assert.False(t, info.StreamStatus.IsNormalEnd())
+}
+
+func TestStreamScannerHandler_StreamStatus_SoftErrors(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(10)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
+		sr.Error(fmt.Errorf("soft error for chunk"))
+	})
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonDone, info.StreamStatus.EndReason)
+	assert.True(t, info.StreamStatus.HasErrors())
+	assert.Equal(t, 10, info.StreamStatus.TotalErrorCount())
+}
+
+func TestStreamScannerHandler_StreamStatus_MultipleErrorsPerChunk(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(5)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
+		sr.Error(fmt.Errorf("error A"))
+		sr.Error(fmt.Errorf("error B"))
+	})
+
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonDone, info.StreamStatus.EndReason)
+	assert.Equal(t, 10, info.StreamStatus.TotalErrorCount())
+}
+
+func TestStreamScannerHandler_StreamStatus_ErrorThenStop(t *testing.T) {
+	t.Parallel()
+
+	// Use a large body without [DONE] to avoid race between scanner's [DONE]
+	// and handler's Stop on the sync.Once EndReason.
+	var b strings.Builder
+	for i := 0; i < 100; i++ {
+		fmt.Fprintf(&b, "data: {\"id\":%d}\n", i)
+	}
+	c, resp, info := setupStreamTest(t, strings.NewReader(b.String()))
+
+	var count atomic.Int64
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
+		count.Add(1)
+		sr.Error(fmt.Errorf("soft error"))
+		sr.Stop(fmt.Errorf("fatal"))
+	})
+
+	assert.Equal(t, int64(1), count.Load())
+	require.NotNil(t, info.StreamStatus)
+	assert.Equal(t, relaycommon.StreamEndReasonHandlerStop, info.StreamStatus.EndReason)
+	assert.Equal(t, 2, info.StreamStatus.TotalErrorCount())
+}
+
+func TestStreamScannerHandler_StreamStatus_InitializedIfNil(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(1)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	assert.Nil(t, info.StreamStatus)
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {})
+
+	assert.NotNil(t, info.StreamStatus)
+}
+
+func TestStreamScannerHandler_StreamStatus_PreInitialized(t *testing.T) {
+	t.Parallel()
+
+	body := buildSSEBody(5)
+	c, resp, info := setupStreamTest(t, strings.NewReader(body))
+
+	info.StreamStatus = relaycommon.NewStreamStatus()
+	info.StreamStatus.RecordError("pre-existing error")
+
+	StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {})
+
+	assert.Equal(t, relaycommon.StreamEndReasonDone, info.StreamStatus.EndReason)
+	assert.Equal(t, 1, info.StreamStatus.TotalErrorCount())
+}
+
 func TestStreamScannerHandler_PingInterleavesWithSlowUpstream(t *testing.T) {
 	t.Parallel()
 
@@ -469,9 +642,6 @@ func TestStreamScannerHandler_PingInterleavesWithSlowUpstream(t *testing.T) {
 		setting.PingIntervalSeconds = oldSeconds
 	})
 
-	// Slow upstream + slow handler. Total stream takes ~5 seconds.
-	// The ping goroutine stays alive as long as the scanner is reading,
-	// so pings should fire between data writes.
 	pr, pw := io.Pipe()
 	go func() {
 		defer pw.Close()
@@ -498,9 +668,8 @@ func TestStreamScannerHandler_PingInterleavesWithSlowUpstream(t *testing.T) {
 	var count atomic.Int64
 	done := make(chan struct{})
 	go func() {
-		StreamScannerHandler(c, resp, info, func(data string) bool {
+		StreamScannerHandler(c, resp, info, func(data string, sr *StreamResult) {
 			count.Add(1)
-			return true
 		})
 		close(done)
 	}()

relay/image_handler.go

@@ -117,11 +117,20 @@ func ImageHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 	if request.N != nil {
 		imageN = *request.N
 	}
+
+	// n is handled via OtherRatio so it is applied exactly once in quota
+	// calculation (both price-based and ratio-based paths).
+	// Adaptors may have already set a more accurate count from the
+	// upstream response; only set the default when they haven't.
+	if _, hasN := info.PriceData.OtherRatios["n"]; !hasN {
+		info.PriceData.AddOtherRatio("n", float64(imageN))
+	}
+
 	if usage.(*dto.Usage).TotalTokens == 0 {
-		usage.(*dto.Usage).TotalTokens = int(imageN)
+		usage.(*dto.Usage).TotalTokens = 1
 	}
 	if usage.(*dto.Usage).PromptTokens == 0 {
-		usage.(*dto.Usage).PromptTokens = int(imageN)
+		usage.(*dto.Usage).PromptTokens = 1
 	}
 
 	quality := "standard"
@@ -141,6 +150,6 @@ func ImageHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *type
 		logContent = append(logContent, fmt.Sprintf("生成数量 %d", imageN))
 	}
 
-	postConsumeQuota(c, info, usage.(*dto.Usage), logContent...)
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), logContent)
 	return nil
 }

relay/mjproxy_handler.go

@@ -49,6 +49,13 @@ func RelayMidjourneyImage(c *gin.Context) {
 	if httpClient == nil {
 		httpClient = service.GetHttpClient()
 	}
+	fetchSetting := system_setting.GetFetchSetting()
+	if err := common.ValidateURLWithFetchSetting(midjourneyTask.ImageUrl, fetchSetting.EnableSSRFProtection, fetchSetting.AllowPrivateIp, fetchSetting.DomainFilterMode, fetchSetting.IpFilterMode, fetchSetting.DomainList, fetchSetting.IpList, fetchSetting.AllowedPorts, fetchSetting.ApplyIPFilterForDomain); err != nil {
+		c.JSON(http.StatusForbidden, gin.H{
+			"error": fmt.Sprintf("request blocked: %v", err),
+		})
+		return
+	}
 	resp, err := httpClient.Get(midjourneyTask.ImageUrl)
 	if err != nil {
 		c.JSON(http.StatusInternalServerError, gin.H{

relay/rerank_handler.go

@@ -96,6 +96,6 @@ func RerankHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *typ
 		service.ResetStatusCode(newAPIError, statusCodeMappingStr)
 		return newAPIError
 	}
-	postConsumeQuota(c, info, usage.(*dto.Usage))
+	service.PostTextConsumeQuota(c, info, usage.(*dto.Usage), nil)
 	return nil
 }

relay/responses_handler.go

@@ -145,7 +145,7 @@ func ResponsesHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 			info.PriceData = originPriceData
 			return types.NewError(err, types.ErrorCodeModelPriceError, types.ErrOptionWithSkipRetry())
 		}
-		postConsumeQuota(c, info, usageDto)
+		service.PostTextConsumeQuota(c, info, usageDto, nil)
 
 		info.OriginModelName = originModelName
 		info.PriceData = originPriceData
@@ -155,7 +155,7 @@ func ResponsesHelper(c *gin.Context, info *relaycommon.RelayInfo) (newAPIError *
 	if strings.HasPrefix(info.OriginModelName, "gpt-4o-audio") {
 		service.PostAudioConsumeQuota(c, info, usageDto, "")
 	} else {
-		postConsumeQuota(c, info, usageDto)
+		service.PostTextConsumeQuota(c, info, usageDto, nil)
 	}
 	return nil
 }

router/api-router.go

@@ -36,10 +36,10 @@ func SetApiRouter(router *gin.Engine) {
 		apiRouter.POST("/user/reset", middleware.CriticalRateLimit(), controller.ResetPassword)
 		// OAuth routes - specific routes must come before :provider wildcard
 		apiRouter.GET("/oauth/state", middleware.CriticalRateLimit(), controller.GenerateOAuthCode)
-		apiRouter.GET("/oauth/email/bind", middleware.CriticalRateLimit(), controller.EmailBind)
+		apiRouter.POST("/oauth/email/bind", middleware.CriticalRateLimit(), controller.EmailBind)
 		// Non-standard OAuth (WeChat, Telegram) - keep original routes
 		apiRouter.GET("/oauth/wechat", middleware.CriticalRateLimit(), controller.WeChatAuth)
-		apiRouter.GET("/oauth/wechat/bind", middleware.CriticalRateLimit(), controller.WeChatBind)
+		apiRouter.POST("/oauth/wechat/bind", middleware.CriticalRateLimit(), controller.WeChatBind)
 		apiRouter.GET("/oauth/telegram/login", middleware.CriticalRateLimit(), controller.TelegramLogin)
 		apiRouter.GET("/oauth/telegram/bind", middleware.CriticalRateLimit(), controller.TelegramBind)
 		// Standard OAuth providers (GitHub, Discord, OIDC, LinuxDO) - unified route
@@ -194,6 +194,8 @@ func SetApiRouter(router *gin.Engine) {
 			performanceRoute.DELETE("/disk_cache", controller.ClearDiskCache)
 			performanceRoute.POST("/reset_stats", controller.ResetPerformanceStats)
 			performanceRoute.POST("/gc", controller.ForceGC)
+			performanceRoute.GET("/logs", controller.GetLogFiles)
+			performanceRoute.DELETE("/logs", controller.CleanupLogFiles)
 		}
 		ratioSyncRoute := apiRouter.Group("/ratio_sync")
 		ratioSyncRoute.Use(middleware.RootAuth())
@@ -224,7 +226,7 @@ func SetApiRouter(router *gin.Engine) {
 			channelRoute.POST("/batch", controller.DeleteChannelBatch)
 			channelRoute.POST("/fix", controller.FixChannelsAbilities)
 			channelRoute.GET("/fetch_models/:id", controller.FetchUpstreamModels)
-			channelRoute.POST("/fetch_models", controller.FetchModels)
+			channelRoute.POST("/fetch_models", middleware.RootAuth(), controller.FetchModels)
 			channelRoute.POST("/codex/oauth/start", controller.StartCodexOAuth)
 			channelRoute.POST("/codex/oauth/complete", controller.CompleteCodexOAuth)
 			channelRoute.POST("/:id/codex/oauth/start", controller.StartCodexOAuthForChannel)
@@ -255,6 +257,7 @@ func SetApiRouter(router *gin.Engine) {
 			tokenRoute.PUT("/", controller.UpdateToken)
 			tokenRoute.DELETE("/:id", controller.DeleteToken)
 			tokenRoute.POST("/batch", controller.DeleteTokenBatch)
+			tokenRoute.POST("/batch/keys", middleware.CriticalRateLimit(), middleware.DisableCache(), controller.GetTokenKeysBatch)
 		}
 
 		usageRoute := apiRouter.Group("/usage")
@@ -290,6 +293,7 @@ func SetApiRouter(router *gin.Engine) {
 
 		dataRoute := apiRouter.Group("/data")
 		dataRoute.GET("/", middleware.AdminAuth(), controller.GetAllQuotaDates)
+		dataRoute.GET("/users", middleware.AdminAuth(), controller.GetQuotaDatesByUser)
 		dataRoute.GET("/self", middleware.UserAuth(), controller.GetUserQuotaDates)
 
 		logRoute.Use(middleware.CORS(), middleware.CriticalRateLimit())

service/channel_affinity.go

@@ -166,12 +166,22 @@ func GetChannelAffinityCacheStats() ChannelAffinityCacheStats {
 			unknown++
 			continue
 		}
-		if rule.IncludeUsingGroup {
+		if rule.IncludeModelName {
 			if len(parts) < 3 {
 				unknown++
 				continue
 			}
 		}
+		if rule.IncludeUsingGroup {
+			minParts := 3
+			if rule.IncludeModelName {
+				minParts = 4
+			}
+			if len(parts) < minParts {
+				unknown++
+				continue
+			}
+		}
 		byRuleName[ruleName]++
 	}
 
@@ -319,11 +329,14 @@ func extractChannelAffinityValue(c *gin.Context, src operation_setting.ChannelAf
 	}
 }
 
-func buildChannelAffinityCacheKeySuffix(rule operation_setting.ChannelAffinityRule, usingGroup string, affinityValue string) string {
-	parts := make([]string, 0, 3)
+func buildChannelAffinityCacheKeySuffix(rule operation_setting.ChannelAffinityRule, modelName string, usingGroup string, affinityValue string) string {
+	parts := make([]string, 0, 4)
 	if rule.IncludeRuleName && rule.Name != "" {
 		parts = append(parts, rule.Name)
 	}
+	if rule.IncludeModelName && modelName != "" {
+		parts = append(parts, modelName)
+	}
 	if rule.IncludeUsingGroup && usingGroup != "" {
 		parts = append(parts, usingGroup)
 	}
@@ -573,7 +586,7 @@ func GetPreferredChannelByAffinity(c *gin.Context, modelName string, usingGroup
 		if ttlSeconds <= 0 {
 			ttlSeconds = setting.DefaultTTLSeconds
 		}
-		cacheKeySuffix := buildChannelAffinityCacheKeySuffix(rule, usingGroup, affinityValue)
+		cacheKeySuffix := buildChannelAffinityCacheKeySuffix(rule, modelName, usingGroup, affinityValue)
 		cacheKeyFull := channelAffinityCacheNamespace + ":" + cacheKeySuffix
 		setChannelAffinityContext(c, channelAffinityMeta{
 			CacheKey:       cacheKeyFull,
@@ -610,14 +623,17 @@ func ShouldSkipRetryAfterChannelAffinityFailure(c *gin.Context) bool {
 		return false
 	}
 	v, ok := c.Get(ginKeyChannelAffinitySkipRetry)
+	if ok {
+		b, ok := v.(bool)
+		if ok {
+			return b
+		}
+	}
+	meta, ok := getChannelAffinityMeta(c)
 	if !ok {
 		return false
 	}
-	b, ok := v.(bool)
-	if !ok {
-		return false
-	}
-	return b
+	return meta.SkipRetry
 }
 
 func MarkChannelAffinityUsed(c *gin.Context, selectedGroup string, channelID int) {