変更履歴

すべての履歴バージョンを表示するには、GitHub Releases ページをご覧ください。このページは定期的に最新の更新情報を取得します。

修复GetUserSetting永远无法读取旧版本用户设置的问题

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.7...v0.11.0-alpha.8

Download Resources

• Source code (zip)
• Source code (tar.gz)

第三方登录支持优先使用第三方用户名 请求头透传功能优化 加强渠道模型选择防呆设计

Third-party login support, prioritize using third-party username Request header pass-through function optimization Strengthen channel model selection foolproof design

What's Changed

• fix: skip Accept-Encoding during header passthrough (#2214) by @Caisin in https://github.com/QuantumNous/new-api/pull/2997
• feat: move user bindings to dedicated management modal by @seefs001 in https://github.com/QuantumNous/new-api/pull/2993
• Fix/pr 2900 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2998
• feat: 支持优先使用第三方用户名 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2900
• feat(web): add custom-model create hint and i18n translations by @RedwindA in https://github.com/QuantumNous/new-api/pull/3002

New Contributors

• @Caisin made their first contribution in https://github.com/QuantumNous/new-api/pull/2997

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.6...v0.11.0-alpha.7

Download Resources

• Source code (zip)
• Source code (tar.gz)

grok 违规扣费完善

What's Changed

• fix: violation fee check by @seefs001 in https://github.com/QuantumNous/new-api/pull/2994

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.5...v0.11.0-alpha.6

Download Resources

• Source code (zip)
• Source code (tar.gz)

修复PostForm下重试出现multipart: NextPart: EOF错误

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.4...v0.11.0-alpha.5

Download Resources

• Source code (zip)
• Source code (tar.gz)

What's Changed

• fix: claude affinity cache counter by @seefs001 in https://github.com/QuantumNous/new-api/pull/2980
• feat: add missing OpenAI/Claude/Gemini request fields by @seefs001 in https://github.com/QuantumNous/new-api/pull/2971
• feat: codex oauth proxy by @seefs001 in https://github.com/QuantumNous/new-api/pull/2961
• feat: minimax native /v1/messages by @seefs001 in https://github.com/QuantumNous/new-api/pull/2960
• fix: unify usage mapping and include toolUsePromptTokenCount by @seefs001 in https://github.com/QuantumNous/new-api/pull/2959

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.3...v0.11.0-alpha.4

Download Resources

• Source code (zip)
• Source code (tar.gz)

加强状态码重定向防呆设计 修复部分异步任务渠道无法获取任务状态的问题

Strengthen foolproof design for status code redirection. Fix issues where some asynchronous task channels cannot obtain task status.

What's Changed

• Feature/channel retry warning by @seefs001 in https://github.com/QuantumNous/new-api/pull/2987

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.2...v0.11.0-alpha.3

Download Resources

• Source code (zip)
• Source code (tar.gz)

完善异步任务超时退款机制，对已存在的旧任务，超时后进行仅失败处理，对于新建任务，增加超时退款流程，超时时间通过环境变量TASK_TIMEOUT_MINUTES设置，默认24小时。

Improve the asynchronous task timeout refund mechanism. For existing old tasks, perform only failure processing after timeout. For newly created tasks, add a timeout refund process. The timeout duration is set via the environment variable TASK_TIMEOUT_MINUTES, defaulting to 24 hours.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0-alpha.1...v0.11.0-alpha.2

Download Resources

• Source code (zip)
• Source code (tar.gz)

重构异步任务逻辑，支持异步任务模型重定向 修复Veo视频模型因id过长生成失败问题 异步任务支持订阅付费模式 使用日志新增退款类型，方便筛选

Refactor asynchronous task logic, support asynchronous task model mapped Fix veo video model generation failure due to excessively long id Add refund type using logs

What's Changed

• refactor: async task by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2985

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9...v0.11.0-alpha.1

Download Resources

• Source code (zip)
• Source code (tar.gz)

新增自定义OAuth提供商功能 Support custom OAuth provider

What's Changed

• feat: Force beta=true parameter for Anthropic channel by @MUTED64 in https://github.com/QuantumNous/new-api/pull/2874
• feat(oauth): implement custom OAuth provider by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2857
• fix: Claude stream block index/type transitions by @seefs001 in https://github.com/QuantumNous/new-api/pull/2854
• fix: add paragraph breaks between reasoning summary chunks by @seefs001 in https://github.com/QuantumNous/new-api/pull/2864
• feat: channel test with stream=true by @seefs001 in https://github.com/QuantumNous/new-api/pull/2875
• fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2883
• fix: 补全 streaming message_delta 事件缺失的 input_tokens 和 cache 相关字段 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2887
• fix: 补全 streaming message_delta 事件缺失的 input_tokens 和 cache 相关字段 (#2881) by @naaive in https://github.com/QuantumNous/new-api/pull/2885
• feat: /v1/messages -> /v1/responses by @seefs001 in https://github.com/QuantumNous/new-api/pull/2889
• feat: make 5m cache-creation ratio configurable by @seefs001 in https://github.com/QuantumNous/new-api/pull/2870
• fix: 如果模型管理有自定义配置则不合并默认配置 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2895
• fix: 修复模型管理"参与官方同步"与"状态"开关无法保存的问题 by @wans10 in https://github.com/QuantumNous/new-api/pull/2815
• 改变端点映射文案 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2896
• optimize： channel affinity tips by @seefs001 in https://github.com/QuantumNous/new-api/pull/2898
• fix streamSupportedChannels by @seefs001 in https://github.com/QuantumNous/new-api/pull/2909
• chore(deps): bump axios from 1.12.0 to 1.13.5 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2917
• feat(localization): added zh_TW by @olivertzeng in https://github.com/QuantumNous/new-api/pull/2913
• feat: logs cache field by @seefs001 in https://github.com/QuantumNous/new-api/pull/2920
• feat(xai): 为xAI渠道添加/v1/responses支持 by @funkpopo in https://github.com/QuantumNous/new-api/pull/2897
• feat: add OpenRouter pricing support to upstream ratio sync by @RedwindA in https://github.com/QuantumNous/new-api/pull/2925
• feat(user): add currency amount input with auto quota conversion by @worryzyy in https://github.com/QuantumNous/new-api/pull/2916
• 优化: 任务日志查询速度并显示用户详情 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2905
• fix(token-search): use TrimPrefix for sk- token normalization by @RedwindA in https://github.com/QuantumNous/new-api/pull/2928
• fix: support numeric status code mapping in ResetStatusCode by @seefs001 in https://github.com/QuantumNous/new-api/pull/2926
• feat(aws): add claude-sonnet-4-6 Bedrock mapping and cross-region support by @wellsgz in https://github.com/QuantumNous/new-api/pull/2976
• feat(ratio-sync): support models.dev ratio sync and fix Gemini cache ratios by @RedwindA in https://github.com/QuantumNous/new-api/pull/2973
• fix: normalize input_text content blocks in Claude-to-OpenAI conversion by @0-don in https://github.com/QuantumNous/new-api/pull/2968
• imporve oauth provider UI/UX by @seefs001 in https://github.com/QuantumNous/new-api/pull/2983

New Contributors

• @MUTED64 made their first contribution in https://github.com/QuantumNous/new-api/pull/2874
• @naaive made their first contribution in https://github.com/QuantumNous/new-api/pull/2885
• @olivertzeng made their first contribution in https://github.com/QuantumNous/new-api/pull/2913
• @funkpopo made their first contribution in https://github.com/QuantumNous/new-api/pull/2897
• @worryzyy made their first contribution in https://github.com/QuantumNous/new-api/pull/2916
• @wellsgz made their first contribution in https://github.com/QuantumNous/new-api/pull/2976
• @0-don made their first contribution in https://github.com/QuantumNous/new-api/pull/2968

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8...v0.10.9

Download Resources

• New-API-App.0.10.9.exe (96.4 MB)
• New-API-App.Setup.0.10.9.exe (96.6 MB)
• new-api-arm64-v0.10.9 (62.1 MB)
• new-api-macos-v0.10.9 (80.1 MB)
• new-api-v0.10.9 (64.4 MB)
• new-api-v0.10.9.exe (66.0 MB)
• Source code (zip)
• Source code (tar.gz)

• Fix the issue of custom OAuth provider callback 404.
• 修复自定义Oauth提供商回调404的问题

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.6...v0.10.9-alpha.7

Download Resources

• Source code (zip)
• Source code (tar.gz)

What's Changed

• feat: add OpenRouter pricing support to upstream ratio sync by @RedwindA in https://github.com/QuantumNous/new-api/pull/2925
• feat(user): add currency amount input with auto quota conversion by @worryzyy in https://github.com/QuantumNous/new-api/pull/2916
• 优化: 任务日志查询速度并显示用户详情 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2905
• fix(token-search): use TrimPrefix for sk- token normalization by @RedwindA in https://github.com/QuantumNous/new-api/pull/2928
• fix: support numeric status code mapping in ResetStatusCode by @seefs001 in https://github.com/QuantumNous/new-api/pull/2926

New Contributors

• @worryzyy made their first contribution in https://github.com/QuantumNous/new-api/pull/2916

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.5...v0.10.9-alpha.6

Download Resources

• Source code (zip)
• Source code (tar.gz)

Significantly optimize memory consumption with disk cache enabled xAI channel add/v1/responses support Optimize cached token log display

大幅优化磁盘缓存开启下的内存消耗，进一步节省内存 xAI渠道添加/v1/responses支持 优化缓存token日志显示

What's Changed

• chore(deps): bump axios from 1.12.0 to 1.13.5 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2917
• feat(localization): added zh_TW by @olivertzeng in https://github.com/QuantumNous/new-api/pull/2913
• feat: logs cache field by @seefs001 in https://github.com/QuantumNous/new-api/pull/2920
• feat(xai): 为xAI渠道添加/v1/responses支持 by @funkpopo in https://github.com/QuantumNous/new-api/pull/2897

New Contributors

• @olivertzeng made their first contribution in https://github.com/QuantumNous/new-api/pull/2913
• @funkpopo made their first contribution in https://github.com/QuantumNous/new-api/pull/2897

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.4...v0.10.9-alpha.5

Download Resources

• Source code (zip)
• Source code (tar.gz)

修复moonshot kimi渠道计费异常的问题

What's Changed

• fix streamSupportedChannels by @seefs001 in https://github.com/QuantumNous/new-api/pull/2909

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.3...v0.10.9-alpha.4

Download Resources

• Source code (zip)
• Source code (tar.gz)

What's Changed

• 改变端点映射文案 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2896
• optimize： channel affinity tips by @seefs001 in https://github.com/QuantumNous/new-api/pull/2898

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.2...v0.10.9-alpha.3

Download Resources

• Source code (zip)
• Source code (tar.gz)

What's Changed

• fix: 如果模型管理有自定义配置则不合并默认配置 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2895
• fix: 修复模型管理"参与官方同步"与"状态"开关无法保存的问题 by @wans10 in https://github.com/QuantumNous/new-api/pull/2815

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.9-alpha.1...v0.10.9-alpha.2

Download Resources

• Source code (zip)
• Source code (tar.gz)

• Support for custom OAuth providers
• 支持自定义Oauth提供商
• 渠道测试支持流测试
• 支持配置缓存创建倍率

What's Changed

• feat: Force beta=true parameter for Anthropic channel by @MUTED64 in https://github.com/QuantumNous/new-api/pull/2874
• feat(oauth): implement custom OAuth provider by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2857
• fix: Claude stream block index/type transitions by @seefs001 in https://github.com/QuantumNous/new-api/pull/2854
• fix: add paragraph breaks between reasoning summary chunks by @seefs001 in https://github.com/QuantumNous/new-api/pull/2864
• feat: channel test with stream=true by @seefs001 in https://github.com/QuantumNous/new-api/pull/2875
• fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2883
• fix: 补全 streaming message_delta 事件缺失的 input_tokens 和 cache 相关字段 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2887
• fix: 补全 streaming message_delta 事件缺失的 input_tokens 和 cache 相关字段 (#2881) by @naaive in https://github.com/QuantumNous/new-api/pull/2885
• feat: /v1/messages -> /v1/responses by @seefs001 in https://github.com/QuantumNous/new-api/pull/2889
• feat: make 5m cache-creation ratio configurable by @seefs001 in https://github.com/QuantumNous/new-api/pull/2870

New Contributors

• @MUTED64 made their first contribution in https://github.com/QuantumNous/new-api/pull/2874
• @naaive made their first contribution in https://github.com/QuantumNous/new-api/pull/2885

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8...v0.10.9-alpha.1

Download Resources

• Source code (zip)
• Source code (tar.gz)

• Paid Subscription Support (Beta): Added support for a paid subscription model.

• System Performance Monitoring: Introduced system performance monitoring. When enabled, the system will reject new Relay requests (e.g., /v1, /v1beta) if resource usage exceeds configured thresholds, ensuring system stability.

• Cache Optimization: Further optimized caching functionality to reduce memory usage during large request bodies.

• Bug Fix: Fixed an issue where Chat2Responses could not retrieve the chain of thought.

• Header Pass-through Support: Added support for full request header pass-through based on Go Regex rules. Please refer to the PR description for usage details: https://github.com/QuantumNous/new-api/pull/2840

• Multi-language Backend Support: Added backend support for multiple languages, respecting user-level language preferences and the Accept-Language request header.

• Task Logs Update: Task logs now display the username.

• Gemini Billing Update: Implemented billing for empty completions caused by Gemini content safety filters.

• Codex Channel Support: Improved support for Codex channels.

• Model Removal: Removed support for claude-2 and claude-1 series models.

• 支持订阅付费模式（测试版）

• 新增系统性能监控，启用性能监控后，当系统资源使用率超过设定阈值时，将拒绝新的 Relay 请求 (/v1, /v1beta 等)，以保护系统稳定性。

• 进一步优化缓存功能，减少大请求体下内存占用

• 修复Chat2Responses无法获取思维链的问题

• 支持基于Go Regex规则和全量的请求头透传，使用方式请参考PR描述 https://github.com/QuantumNous/new-api/pull/2840

• 支持后端多语言，支持用户级别语言偏好和Accept-Language请求头偏好

• 任务日志显示用户名

• 对Gemini因审核导致的空补全进行扣费处理

• 完善Codex渠道支持

• 移除对claude-2和claude-1系列模型支持

What's Changed

• feat: add regex pattern to mask API keys in sensitive information by @xiangsx in https://github.com/QuantumNous/new-api/pull/2587
• fix(task): 修复使用 auto 分组时 Task Relay 不记录日志和不扣费的问题 by @147API in https://github.com/QuantumNous/new-api/pull/2588
• fix: 设置默认max req body 为128MB by @xyfacai in https://github.com/QuantumNous/new-api/pull/2590
• fix: remove Minimax from FETCHABLE channels by @RedwindA in https://github.com/QuantumNous/new-api/pull/2619
• fix(gemini): fetch model list via native v1beta/models endpoint by @RedwindA in https://github.com/QuantumNous/new-api/pull/2615
• feat: /v1/chat/completion -> /v1/response by @seefs001 in https://github.com/QuantumNous/new-api/pull/2629
• fix: support snake_case fields in GeminiChatGenerationConfig by @deanxv in https://github.com/QuantumNous/new-api/pull/2646
• feat: status code auto-disable configuration by @seefs001 in https://github.com/QuantumNous/new-api/pull/2647
• fix: chat2response setting ui by @seefs001 in https://github.com/QuantumNous/new-api/pull/2643
• fix: clean propertyNames for gemini function by @seefs001 in https://github.com/QuantumNous/new-api/pull/2642
• feat: channel testing supports parameter overriding by @seefs001 in https://github.com/QuantumNous/new-api/pull/2627
• 新增: 豆包视频1.5pro by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2632
• feat: codex channel by @seefs001 in https://github.com/QuantumNous/new-api/pull/2652
• fix: log info input_token (chat -> claude model ) by @seefs001 in https://github.com/QuantumNous/new-api/pull/2670
• fix: update warning threshold label from '5$' to '2$' by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2690
• fix: fix email send when special character(./+) included in email string by @Bliod-Cook in https://github.com/QuantumNous/new-api/pull/2696
• fix: video content api Priority use url field by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2691
• fix: jimeng i2v support multi image by metadata by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2688
• Revert "fix: video content api Priority use url field" by @seefs001 in https://github.com/QuantumNous/new-api/pull/2710
• fix: openAI function to gemini function field adjusted to whitelist mode by @seefs001 in https://github.com/QuantumNous/new-api/pull/2667
• feat: TLS_INSECURE_SKIP_VERIFY env by @seefs001 in https://github.com/QuantumNous/new-api/pull/2668
• fix: the login method cannot be displayed under the aff link. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2676
• fix: codex Unsupported parameter: max_output_tokens by @seefs001 in https://github.com/QuantumNous/new-api/pull/2684
• feat: customizable automatic retry status codes by @seefs001 in https://github.com/QuantumNous/new-api/pull/2663
• fix: calls to multiple tools in gemini all return index=0 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2701
• fix: replace Alibaba's Claude-compatible url with new url by @seefs001 in https://github.com/QuantumNous/new-api/pull/2702
• feat: log shows request conversion by @seefs001 in https://github.com/QuantumNous/new-api/pull/2703
• feat(qwen): support qwen image sync image model config by @xyfacai in https://github.com/QuantumNous/new-api/pull/2717
• feat: requestId time string use UTC by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2713
• fix request pass-through aws channels can't test by @daggeryu in https://github.com/QuantumNous/new-api/pull/2693
• feat: enhance Authorization header handling with Header Override support by @Li-Xingyu in https://github.com/QuantumNous/new-api/pull/2738
• feat(gemini): 支持 tool_choice 参数转换，优化多个渠道错误处理 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2742
• 支持 tool_choice 参数转换，优化错误处理 by @laozuzhen in https://github.com/QuantumNous/new-api/pull/2540
• 修复钱包页面刷新时显示错误余额数据的问题 by @wans10 in https://github.com/QuantumNous/new-api/pull/2749
• feat: 可视化模型重定向设置中，允许用户从拉取的模型列表里直接选择模型 by @Bliod-Cook in https://github.com/QuantumNous/new-api/pull/2610
• feat: header passthrough by @seefs001 in https://github.com/QuantumNous/new-api/pull/2735
• fix: test using the correct path for rerank by @seefs001 in https://github.com/QuantumNous/new-api/pull/2736
• feat: channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2669
• feat: openai response /v1/response/compact by @seefs001 in https://github.com/QuantumNous/new-api/pull/2644
• feat: grok Usage Guidelines Violation Fee by @seefs001 in https://github.com/QuantumNous/new-api/pull/2753
• feat(gemini): map OpenAI stop to Gemini stopSequences by @RedwindA in https://github.com/QuantumNous/new-api/pull/2779
• fix: remove disable_parallel_tool_use if tool_choice=none by @seefs001 in https://github.com/QuantumNous/new-api/pull/2765
• fix: /v1/responses/compact default billing by @seefs001 in https://github.com/QuantumNous/new-api/pull/2766
• feat: disk request body cache by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2780
• feat: Support customizing the success and cancel url of Stripe. by @mehunk in https://github.com/QuantumNous/new-api/pull/2745
• feat: doubao add first and last image to video by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2756
• feat: task pre consume modelPrice default use setting value by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2733
• feat: CodeViewer click link and auto wrap by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2783
• feat(gemini): support cached token billing by @RedwindA in https://github.com/QuantumNous/new-api/pull/2798
• fix(ui): use distinct color palette for group tags by @KiGamji in https://github.com/QuantumNous/new-api/pull/2759
• fix: channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2799
• fix: make channel Host override take effect by @lanfunoe in https://github.com/QuantumNous/new-api/pull/2805
• fix: claude panic by @seefs001 in https://github.com/QuantumNous/new-api/pull/2804
• feat: /v1/responses qwen3 max && perplexity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2803
• feat: auto-adapt video modal by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2764
• fix: openrouter claude cache usage by @seefs001 in https://github.com/QuantumNous/new-api/pull/2811
• feat：default enable channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2809
• fix: vertex maas api addr by @seefs001 in https://github.com/QuantumNous/new-api/pull/2810
• ✨ feat: add subscription billing system by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2808
• fix: 移除不必要的 unescapeMapOrSlice 调用，修复 Windows 路径转义问题 by @thirking in https://github.com/QuantumNous/new-api/pull/2814
• feat: add useTimeSeconds in error log by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2819
• feat: log search field request_id && conversion_path display by @seefs001 in https://github.com/QuantumNous/new-api/pull/2825
• Revert "fix(ui): use distinct color palette for group tags" by @seefs001 in https://github.com/QuantumNous/new-api/pull/2832
• feat(performance): implement system performance monitoring by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2835
• feat: task log show username by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2635
• ✨ fix: Improve subscription payment handling and card layout consistency by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2838
• 🐛 fix: sidebar scroll on mobile dynamic viewport by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2839
• fix: map Responses reasoning stream to chat completion deltas by @seefs001 in https://github.com/QuantumNous/new-api/pull/2837
• feat: 支持基于Go Regex规则和全量的请求体透传 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2840
• feat: backend i18n by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2842
• fix: optimize Codex relay by @dahetaoa in https://github.com/QuantumNous/new-api/pull/2826
• fix: charge local input tokens when Gemini returns empty response by @seefs001 in https://github.com/QuantumNous/new-api/pull/2848
• remove: drop support for claude-2 and claude-1 series models by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2853
• feat: add claude-opus-4-6 by @prnake in https://github.com/QuantumNous/new-api/pull/2863
• fix: /v1/chat/completions -> /v1/responses json_schema by @seefs001 in https://github.com/QuantumNous/new-api/pull/2876
• ✨ refactor(wallet): Top-up layout to embed subscription plans into the recharge card tabs by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2878
• refactor: 抽象统一计费会话 BillingSession by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2877
• 🔔 feat: Add subscription-aware quota notifications and update UI copy by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2880
• ✨ chore: Improve subscription billing fallback and UI states by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2879
• feat: add Codex channel disclaimer (i18n, OpenAI terms) by @seefs001 in https://github.com/QuantumNous/new-api/pull/2852

New Contributors

• @xiangsx made their first contribution in https://github.com/QuantumNous/new-api/pull/2587
• @147API made their first contribution in https://github.com/QuantumNous/new-api/pull/2588
• @deanxv made their first contribution in https://github.com/QuantumNous/new-api/pull/2646
• @Li-Xingyu made their first contribution in https://github.com/QuantumNous/new-api/pull/2738
• @laozuzhen made their first contribution in https://github.com/QuantumNous/new-api/pull/2540
• @mehunk made their first contribution in https://github.com/QuantumNous/new-api/pull/2745
• @KiGamji made their first contribution in https://github.com/QuantumNous/new-api/pull/2759
• @lanfunoe made their first contribution in https://github.com/QuantumNous/new-api/pull/2805
• @thirking made their first contribution in https://github.com/QuantumNous/new-api/pull/2814
• @dahetaoa made their first contribution in https://github.com/QuantumNous/new-api/pull/2826

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.5...v0.10.8

Download Resources

• New-API-App.0.10.8.exe (96.4 MB)
• New-API-App.Setup.0.10.8.exe (96.6 MB)
• new-api-arm64-v0.10.8 (61.8 MB)
• new-api-macos-v0.10.8 (79.7 MB)
• new-api-v0.10.8 (64.1 MB)
• new-api-v0.10.8.exe (65.7 MB)
• Source code (zip)
• Source code (tar.gz)

What's Changed

• feat: add claude-opus-4-6 by @prnake in https://github.com/QuantumNous/new-api/pull/2863

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.8...v0.10.8-alpha.9

Download Resources

• Source code (zip)
• Source code (tar.gz)

• Defaulting to subscriptions when available and avoiding initial flash when no plans exist.
• Adjust the wide-screen layout to place wallet and invite sections side by side, simplify the subscription header and controls, and add padding to prevent card borders from clipping.
• Update related i18n strings by adding the new tab label and removing the obsolete subscription blurb.

What's Changed

• fix: /v1/chat/completions -> /v1/responses json_schema by @seefs001 in https://github.com/QuantumNous/new-api/pull/2876
• ✨ refactor(wallet): Top-up layout to embed subscription plans into the recharge card tabs by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2878
• refactor: 抽象统一计费会话 BillingSession by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2877

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.11...v0.10.8-alpha.12

Download Resources

• Source code (zip)
• Source code (tar.gz)

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.10...v0.10.8-alpha.11

Download Resources

• Source code (zip)
• Source code (tar.gz)

对Gemini因审核导致的空补全进行扣费处理 完善Codex渠道支持 移除对claude-2和claude-1系列模型支持

What's Changed

• fix: optimize Codex relay by @dahetaoa in https://github.com/QuantumNous/new-api/pull/2826
• fix: charge local input tokens when Gemini returns empty response by @seefs001 in https://github.com/QuantumNous/new-api/pull/2848
• remove: drop support for claude-2 and claude-1 series models by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2853

New Contributors

• @dahetaoa made their first contribution in https://github.com/QuantumNous/new-api/pull/2826

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.7...v0.10.8-alpha.8

Download Resources

• Source code (zip)
• Source code (tar.gz)

Support backend multi-language, supporting user-level language preference and Accept-Language request header preference Fix subscription package payment issue Task log displays username Fixed the issue where Chat2Responses could not retrieve the Chain of Thought. Added support for full request body passthrough based on Go Regex rules. Please refer to PR description #2840 for usage.

支持后端多语言，支持用户级别语言偏好和Accept-Language请求头偏好 修复订阅套餐无法支付的问题 任务日志显示用户名 修复Chat2Responses无法获取思维链的问题 支持基于Go Regex规则和全量的请求体透传，使用方式请参考PR描述 #2840

What's Changed

• feat: task log show username by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2635
• ✨ fix: Improve subscription payment handling and card layout consistency by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2838
• 🐛 fix: sidebar scroll on mobile dynamic viewport by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2839
• fix: map Responses reasoning stream to chat completion deltas by @seefs001 in https://github.com/QuantumNous/new-api/pull/2837
• feat: 支持基于Go Regex规则和全量的请求体透传 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2840
• feat: backend i18n by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2842

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.6...v0.10.8-alpha.7

Download Resources

• Source code (zip)
• Source code (tar.gz)

Added system performance monitoring. When performance monitoring is enabled, new Relay requests (/v1, /v1beta, etc.) will be rejected if system resource utilization exceeds the set threshold, in order to protect system stability.

新增系统性能监控，启用性能监控后，当系统资源使用率超过设定阈值时，将拒绝新的 Relay 请求 (/v1, /v1beta 等)，以保护系统稳定性。

What's Changed

• Revert "fix(ui): use distinct color palette for group tags" by @seefs001 in https://github.com/QuantumNous/new-api/pull/2832
• feat(performance): implement system performance monitoring by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2835

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.5...v0.10.8-alpha.6

Download Resources

• Source code (zip)
• Source code (tar.gz)

Log increase request_id search.
Further optimize cache functionality, reducing memory usage for large request bodies.

日志增加request_id搜索
进一步优化缓存功能，减少大请求体下内存占用

What's Changed

• feat: add useTimeSeconds in error log by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2819
• feat: log search field request_id && conversion_path display by @seefs001 in https://github.com/QuantumNous/new-api/pull/2825

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.4...v0.10.8-alpha.5

Download Resources

• Source code (zip)
• Source code (tar.gz)

Try to fix startup issues while using SQLite database 尝试修复sqlite数据库下的启动问题

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.3...v0.10.8-alpha.4

Download Resources

• Source code (zip)
• Source code (tar.gz)

fix epay callback 修复易支付回调出错问题 Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.2...v0.10.8-alpha.3

Download Resources

• Source code (zip)
• Source code (tar.gz)

feat(subscription): validate price amount and migrate database column type 修复pgsql下无法启动的问题

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.8-alpha.1...v0.10.8-alpha.2

Download Resources

• Source code (zip)
• Source code (tar.gz)

Add subscription billing system (beta) 支持订阅付费模式（测试版）

What's Changed

• ✨ feat: add subscription billing system by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2808
• fix: 移除不必要的 unescapeMapOrSlice 调用，修复 Windows 路径转义问题 by @thirking in https://github.com/QuantumNous/new-api/pull/2814

New Contributors

• @thirking made their first contribution in https://github.com/QuantumNous/new-api/pull/2814

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.7...v0.10.8-alpha.1

Download Resources

• Source code (zip)
• Source code (tar.gz)

Major Updates

• Added disk request body caching to System Settings - 性能设置, which can significantly reduce memory usage for large request bodies (swapping memory to disk).
• Supports configuring request header pass-through, see configuration

{
  "Authorization": "{client_header:Test_Header}"
}

• Compatible with Grok's charging for non-compliant requests, with configuration support (System Settings - Model Related Settings - Grok Settings)
• Supports codex context compression/v1/response/compact (requires adding a model, e.g., to use gpt-5.2, you need to configure gpt-5.2-openai-compact)
• Supports channel preference, prioritizing a certain channel based on a field in the request (System Settings - Model Related Settings - Channel Affinity)
• Displays detailed reasons for errors rejected due to policy violations in logs

主要更新

• 系统设置-性能设置 增加磁盘请求体缓存，可大幅减少大请求体下的内存占用（磁盘换内存）
• 支持配置请求头透传，配置参考

{
  "Authorization": "{client_header:Test_Header}"
}

• 兼容Grok对不合规的请求进行收费操作，支持配置 （系统设置 - 模型相关设置 - Grok设置）
• 支持codex上下文压缩/v1/response/compact （需要添加模型，例如要使用gpt-5.2，就需要配置gpt-5.2-openai-compact）
• 支持渠道优选，根据请求某个字段优先选择某个渠道 (系统设置 - 模型相关设置 - 渠道亲和性)
• 日志详情展示由于违反政策被拒绝的错误原因

What's Changed

• feat: add regex pattern to mask API keys in sensitive information by @xiangsx in https://github.com/QuantumNous/new-api/pull/2587
• fix(task): 修复使用 auto 分组时 Task Relay 不记录日志和不扣费的问题 by @147API in https://github.com/QuantumNous/new-api/pull/2588
• fix: 设置默认max req body 为128MB by @xyfacai in https://github.com/QuantumNous/new-api/pull/2590
• fix: remove Minimax from FETCHABLE channels by @RedwindA in https://github.com/QuantumNous/new-api/pull/2619
• fix(gemini): fetch model list via native v1beta/models endpoint by @RedwindA in https://github.com/QuantumNous/new-api/pull/2615
• feat: /v1/chat/completion -> /v1/response by @seefs001 in https://github.com/QuantumNous/new-api/pull/2629
• fix: support snake_case fields in GeminiChatGenerationConfig by @deanxv in https://github.com/QuantumNous/new-api/pull/2646
• feat: status code auto-disable configuration by @seefs001 in https://github.com/QuantumNous/new-api/pull/2647
• fix: chat2response setting ui by @seefs001 in https://github.com/QuantumNous/new-api/pull/2643
• fix: clean propertyNames for gemini function by @seefs001 in https://github.com/QuantumNous/new-api/pull/2642
• feat: channel testing supports parameter overriding by @seefs001 in https://github.com/QuantumNous/new-api/pull/2627
• 新增: 豆包视频1.5pro by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2632
• feat: codex channel by @seefs001 in https://github.com/QuantumNous/new-api/pull/2652
• fix: log info input_token (chat -> claude model ) by @seefs001 in https://github.com/QuantumNous/new-api/pull/2670
• fix: update warning threshold label from '5$' to '2$' by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2690
• fix: fix email send when special character(./+) included in email string by @Bliod-Cook in https://github.com/QuantumNous/new-api/pull/2696
• fix: video content api Priority use url field by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2691
• fix: jimeng i2v support multi image by metadata by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2688
• Revert "fix: video content api Priority use url field" by @seefs001 in https://github.com/QuantumNous/new-api/pull/2710
• fix: openAI function to gemini function field adjusted to whitelist mode by @seefs001 in https://github.com/QuantumNous/new-api/pull/2667
• feat: TLS_INSECURE_SKIP_VERIFY env by @seefs001 in https://github.com/QuantumNous/new-api/pull/2668
• fix: the login method cannot be displayed under the aff link. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2676
• fix: codex Unsupported parameter: max_output_tokens by @seefs001 in https://github.com/QuantumNous/new-api/pull/2684
• feat: customizable automatic retry status codes by @seefs001 in https://github.com/QuantumNous/new-api/pull/2663
• fix: calls to multiple tools in gemini all return index=0 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2701
• fix: replace Alibaba's Claude-compatible url with new url by @seefs001 in https://github.com/QuantumNous/new-api/pull/2702
• feat: log shows request conversion by @seefs001 in https://github.com/QuantumNous/new-api/pull/2703
• feat(qwen): support qwen image sync image model config by @xyfacai in https://github.com/QuantumNous/new-api/pull/2717
• feat: requestId time string use UTC by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2713
• fix request pass-through aws channels can't test by @daggeryu in https://github.com/QuantumNous/new-api/pull/2693
• feat: enhance Authorization header handling with Header Override support by @Li-Xingyu in https://github.com/QuantumNous/new-api/pull/2738
• feat(gemini): 支持 tool_choice 参数转换，优化多个渠道错误处理 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2742
• 支持 tool_choice 参数转换，优化错误处理 by @laozuzhen in https://github.com/QuantumNous/new-api/pull/2540
• 修复钱包页面刷新时显示错误余额数据的问题 by @wans10 in https://github.com/QuantumNous/new-api/pull/2749
• feat: 可视化模型重定向设置中，允许用户从拉取的模型列表里直接选择模型 by @Bliod-Cook in https://github.com/QuantumNous/new-api/pull/2610
• feat: header passthrough by @seefs001 in https://github.com/QuantumNous/new-api/pull/2735
• fix: test using the correct path for rerank by @seefs001 in https://github.com/QuantumNous/new-api/pull/2736
• feat: channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2669
• feat: openai response /v1/response/compact by @seefs001 in https://github.com/QuantumNous/new-api/pull/2644
• feat: grok Usage Guidelines Violation Fee by @seefs001 in https://github.com/QuantumNous/new-api/pull/2753
• feat(gemini): map OpenAI stop to Gemini stopSequences by @RedwindA in https://github.com/QuantumNous/new-api/pull/2779
• fix: remove disable_parallel_tool_use if tool_choice=none by @seefs001 in https://github.com/QuantumNous/new-api/pull/2765
• fix: /v1/responses/compact default billing by @seefs001 in https://github.com/QuantumNous/new-api/pull/2766
• feat: disk request body cache by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2780
• feat: Support customizing the success and cancel url of Stripe. by @mehunk in https://github.com/QuantumNous/new-api/pull/2745
• feat: doubao add first and last image to video by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2756
• feat: task pre consume modelPrice default use setting value by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2733
• feat: CodeViewer click link and auto wrap by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2783
• feat(gemini): support cached token billing by @RedwindA in https://github.com/QuantumNous/new-api/pull/2798
• fix(ui): use distinct color palette for group tags by @KiGamji in https://github.com/QuantumNous/new-api/pull/2759
• fix: channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2799
• fix: make channel Host override take effect by @lanfunoe in https://github.com/QuantumNous/new-api/pull/2805
• fix: claude panic by @seefs001 in https://github.com/QuantumNous/new-api/pull/2804
• feat: /v1/responses qwen3 max && perplexity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2803
• feat: auto-adapt video modal by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2764
• fix: openrouter claude cache usage by @seefs001 in https://github.com/QuantumNous/new-api/pull/2811
• feat：default enable channel affinity by @seefs001 in https://github.com/QuantumNous/new-api/pull/2809
• fix: vertex maas api addr by @seefs001 in https://github.com/QuantumNous/new-api/pull/2810

New Contributors

• @xiangsx made their first contribution in https://github.com/QuantumNous/new-api/pull/2587
• @147API made their first contribution in https://github.com/QuantumNous/new-api/pull/2588
• @deanxv made their first contribution in https://github.com/QuantumNous/new-api/pull/2646
• @Li-Xingyu made their first contribution in https://github.com/QuantumNous/new-api/pull/2738
• @laozuzhen made their first contribution in https://github.com/QuantumNous/new-api/pull/2540
• @mehunk made their first contribution in https://github.com/QuantumNous/new-api/pull/2745
• @KiGamji made their first contribution in https://github.com/QuantumNous/new-api/pull/2759
• @lanfunoe made their first contribution in https://github.com/QuantumNous/new-api/pull/2805

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.5...v0.10.7

Download Resources

• New-API-App.0.10.7.exe (96.2 MB)
• New-API-App.Setup.0.10.7.exe (96.4 MB)
• new-api-arm64-v0.10.7 (60.4 MB)
• new-api-macos-v0.10.7 (77.6 MB)
• new-api-v0.10.7 (62.5 MB)
• new-api-v0.10.7.exe (64.1 MB)
• Source code (zip)
• Source code (tar.gz)

Major Updates

• Added disk request body caching to System Settings - 性能设置, which can significantly reduce memory usage for large request bodies (swapping memory to disk).

主要更新

• 系统设置-性能设置 增加磁盘请求体缓存，可大幅减少大请求体下的内存占用（磁盘换内存）

What's Changed

• feat(gemini): map OpenAI stop to Gemini stopSequences by @RedwindA in https://github.com/QuantumNous/new-api/pull/2779
• fix: remove disable_parallel_tool_use if tool_choice=none by @seefs001 in https://github.com/QuantumNous/new-api/pull/2765
• fix: /v1/responses/compact default billing by @seefs001 in https://github.com/QuantumNous/new-api/pull/2766
• feat: disk request body cache by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2780

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.7-alpha.1...v0.10.7-alpha.2

Download Resources

• Source code (zip)
• Source code (tar.gz)