New API: baixar versions anteriors

What's Changed
- fix(relay): 正确结束已完成的 Responses 流 by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7956
- feat(billing): bill OpenAI 12-hour prompt cache writes at their own price by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7965
- fix(channels): preserve advanced custom body passthrough settings by @seefs001 in https://github.com/QuantumNous/new-api/pull/7971
- fix(logging): 隐藏访问日志中的查询凭据 by @ceezyyy in https://github.com/QuantumNous/new-api/pull/7972
- fix: prevent stale values when switching edit drawer records by @seefs001 in https://github.com/QuantumNous/new-api/pull/7989
- chore(deps): bump react-resizable-panels from 4.12.2 to 4.14.6 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7982
- chore(deps): bump input-otp from 1.4.2 to 1.5.0 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7980
- chore(deps): bump i18next from 26.3.6 to 26.5.3 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7979
- chore(deps): bump react-i18next from 17.0.10 to 17.1.0 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7981
- chore(deps): bump axios from 1.18.1 to 1.20.0 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7835
New Contributors
- @ceezyyy made their first contribution in https://github.com/QuantumNous/new-api/pull/7972
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.46...v1.0.0-rc.47

Important
修复 CVE-2026-111943，请尽快升级。这一版用 Go 1.26.10。Go 校验证书时，会拒绝一小批已经不再可信的中间证书。用源码编译的，Go 也要换成 1.26.10。
Fixes CVE-2026-111943. Please upgrade soon. This release runs on Go 1.26.10. When Go checks a certificate, it rejects a small set of intermediate certificates that are no longer trusted. Builds from source need Go 1.26.10 too.
中文
- 安全：修复 CVE-2026-111943。运行时改为 Go 1.26.10，校验证书时会拒绝一小批已经不再可信的中间证书。
- 邮件：新增用 Resend 发邮件的功能，可以代替 SMTP。发件地址和发件人名称单独填。
- 数据看板：新增缓存用量的显示，和总 token 放在一起。
- 向量：新增阿里云百炼多模态向量模型 qwen3-vl-embedding 的支持，同系列的视觉向量模型也走这个接口。
- 任务：修复任务换渠道重试时还按上一次价格计费的问题。更贵的渠道会先补预扣，余额不够就不会发出去。修复任务提交和轮询没用上「请求头覆盖」的问题。修复轮询跳到别的地址时密钥跟着过去的问题。
- 渠道：修复可灵不接受单个密钥的问题。插件 1.1.1，不带竖线的密钥可以直接用。修复 DeepSeek 美元余额是 0、人民币有钱时，余额显示成 0 的问题。
- 修复：修复通知限流每到整点就重新计数的问题，超过 60 分钟也按你设的时长算。修复 Claude 流式请求在准备和重试期间断开的问题。修复连续的助手消息合并后丢掉工具调用、Claude 报错的问题。修复 Redis 清掉脚本后限流失效的问题。发给 Gemini 的向量请求里带了图片或视频时，会直接拒绝，不再计费。
English
- Security: Fixed CVE-2026-111943 by moving to Go 1.26.10. Certificate checks reject a small set of intermediate certificates that are no longer trusted.
- Email: Added Resend for sending mail, in place of SMTP. The sender address and sender name are set on their own.
- Dashboard: Added cache usage next to the total token count.
- Embeddings: Added Alibaba Bailian multimodal embeddings, starting with qwen3-vl-embedding. The other vision embedding models on Bailian use the same endpoint.
- Tasks: Fixed a retry on another channel still using the previous price. A more expensive channel reserves the difference first, and is not called when the balance cannot cover it. Fixed task submission and polling ignoring Request Header Override. Fixed polling redirects carrying the key to another address.
- Channels: Fixed Kling rejecting a single key. Plugin 1.1.1 accepts a key with no vertical bar. Fixed DeepSeek balance showing 0 when the USD wallet is 0 and the CNY wallet has money.
- Fixes: Fixed notification rate limits resetting at each hour, so a window longer than 60 minutes lasts for the time you set. Fixed Claude streams dropping during preparation and retries. Fixed merged assistant messages dropping tool calls, which made Claude reject the request. Fixed rate limiting after Redis flushed its scripts. An embedding sent to Gemini with an image or video is rejected and not billed.

What's Changed
- fix(notify): keep notification limit window from resetting every hour by @RedwindA in https://github.com/QuantumNous/new-api/pull/7877
- fix(audit): add composite index for audit log ordering by @stl25 in https://github.com/QuantumNous/new-api/pull/7874
- fix(relay): Claude 流式请求在准备和重试期间持续保活 by @mameikagou in https://github.com/QuantumNous/new-api/pull/7857
- feat(email): 添加 resend 邮件服务商接口 by @imysen in https://github.com/QuantumNous/new-api/pull/7880
- feat(dashboard): show cache usage alongside total tokens by @seefs001 in https://github.com/QuantumNous/new-api/pull/7773
- fix(task): 修复任务提交与轮询的请求头覆盖 by @qinhanlei in https://github.com/QuantumNous/new-api/pull/7918
- feat: 支持阿里多模态向量化模型 qwen3-vl-embedding by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5179
- fix(relaykit): keep tool calls when merging adjacent assistant messages for Claude by @rjchien728 in https://github.com/QuantumNous/new-api/pull/7921
- fix(kling): 支持单 API Key by @ZIYUT in https://github.com/QuantumNous/new-api/pull/7897
- fix(limiter): recover redis script after flush(#6412) by @BradleyZhang in https://github.com/QuantumNous/new-api/pull/6486
- fix: fall back to CNY when the DeepSeek USD wallet is zero by @brushax in https://github.com/QuantumNous/new-api/pull/7917
- chore(deps): bump github.com/klauspost/compress from 1.18.3 to 1.18.7 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7931
- chore(deps-dev): bump vitest from 4.1.10 to 5.0.4 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/7838
- fix: upgrade Go to 1.26.10 for CVE-2026-111943 by @seefs001 in https://github.com/QuantumNous/new-api/pull/7942
New Contributors
- @stl25 made their 

Important
v1 正式版即将发布，现在是最后几个 RC 版本。新增 DeepSeek V4.1 Flash、Search 和 Fetch，通知分开开关，以及拖动排列渠道优先级的功能。
敏感词过滤默认关着。以前填过敏感词、但没保存过这个开关的，升级后到「请求检查」里打开。
v1 is coming soon. These are the last few release candidates. Added DeepSeek V4.1 Flash, Search and Fetch, separate notification switches, and dragging to set channel priority.
Sensitive word filtering is off by default. If you filled in a word list and never saved the switch, turn it on under Request checks after upgrading.
中文
- DeepSeek：新增 DeepSeek V4.1 Flash（deepseek-flash）和 deepseek-v4-pro 的支持，后面可以加 -none、-low、-max，@thinking 和 @effort 也能用。新增 Search 和 Fetch，地址是 /v1/tools/search 和 /v1/tools/fetch。Claude Code 和 Codex 让 DeepSeek 去搜索、打开网页时，按次数计费，每 1000 次 0.75 美元。
- 通知：新增按种类开关通知的功能，余额不足提醒默认开着。验证码、重置密码、绑定邮箱，看 SMTP 有没有配好。余额低于你设的数时提醒一次，充回去之后，下次再低才会再提醒。
- 渠道：新增拖动调整渠道优先级的功能，在渠道列表打开「拖拽模式」。新增 New API 渠道查询上游余额的功能，也可以绑定海螺、Vidu 和 Veo。新增令牌「允许查询用户额度」，新建的默认开着，以前建的还是关的。
- 限流和计费：修复分组限流算到一起的问题，某个分组自己设了限流，就按这个分组的次数算。修复 Vidu 收费和实际生成对不上的问题，按它实际生成的模型、时长和分辨率计费。修复 Claude 流式把上游报的 0 丢掉的问题，报了 0 就按 0 算。提示词全在缓存里时，按读缓存的价格计费。
- 报错：修复请求内容读失败时看不出原因的问题。上游报错里，你填的地址、IP 和密钥会被藏起来，文档链接和邮箱还看得到。
- 其他：新增 Sub2API 查询余额的功能。新增 AWS 的 Mantle（Messages API），已经在用的还是 InvokeModel。新增 Vertex 用 API Key 时填写项目 ID 的功能。任务插件这一版更快。
- 修复：修复 Vertex 的 us、eu 访问地址错误的问题。修复容器里过载保护看错 CPU 和内存的问题，看的是容器自己的，连续大约 15 秒都过高才会拒绝请求。
English
- DeepSeek: Added deepseek-flash and deepseek-v4-pro. You can add -none, -low, or -max, and @thinking and @effort work too. Added Search and Fetch at /v1/tools/search and /v1/tools/fetch. When Claude Code or Codex has DeepSeek search or open a page, that is charged per request, $0.75 per 1,000.
- Notifications: Added a switch for each kind of notification. The low balance reminder starts on. Verification codes, password resets, and email binding go out when SMTP is set up. You get one reminder when the balance goes below the number you set, and another the next time it does, after a top-up.
- Channels: Added dragging to change channel priority. Turn on Drag mode in the channel list. Added upstream balance checks for New API channels, which can also bind Hailuo, Vidu, and Veo. Added Allow querying user quota on tokens. New tokens start with it on. Older tokens stay off.
- Rate limits and billing: Fixed group rate limits being counted together. A group with its own limit is counted on its own. Fixed Vidu charges that did not match the video it generated. The charge uses the model, duration, and resolution it actually made. Fixed Claude streams dropping a 0 from the upstream. A reported 0 stays 0. When the whole prompt is cached, the charge is the cache-read price.
- Errors: Fixed request failures that hid the real reason the body could not be read. If it is too large, the response is 413. In an upstream error, the address, IP, and key you entered are hidden. Documentation links and email addresses stay visible.
- Also: Added balance checks for Sub2API channels. Added Mantle (Messages API) for AWS channels. Ones already in use stay on InvokeModel. Added a project ID field for Vertex API Key mode. Task plugins are faster in this release.
- Fixes: Fixed wrong Vertex addresses for the us and eu regions. Fixed overload protection in containers reading the wrong CPU and memory. It uses the container's own, and turns requests away only after they stay too high for about 15 seconds.

What's Changed
- feat(ui): show newly added pricing groups at the top by @seefs001 in https://github.com/QuantumNous/new-api/pull/7782
- feat(channel): support Sub2API balance queries by @AdoreJc in https://github.com/QuantumNous/new-api/pull/7749
- fix: disable sensitive word filtering by default and clear list by @seefs001 in https://github.com/QuantumNous/new-api/pull/7814
- fix(vertex): pass project_id in API Key mode to avoid 404 on official endpoints by @xiaoyuyu6420 in https://github.com/QuantumNous/new-api/pull/6330
- fix(web): sort pricing models by input, output and per-request price by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7813
-

Important
v1 Coming Soon & Log Export / v1 即将发布与日志导出
v1 正式版即将发布，现在是最后几个 RC 版本。
使用日志支持导出 CSV。可以在日志维护里设置是否允许用户导出，以及日志最长时间范围，默认 365 天。普通用户的日志查询、导出和数据看板都按这个范围限制。
游乐场的数据按账号分开保存，升级后之前的聊天记录会清空一次。
Claude 请求按模型支持的档位发送思考强度。发给 Gemini 的工具和结构化输出保留完整的 JSON Schema。
v1 is coming soon. These are the last few release candidates.
Usage logs can be exported as CSV. In Log Maintenance you can choose whether users may export and set the longest log time range, 365 days by default. Regular users' log searches, exports, and dashboard charts all follow that range.
Playground data is stored per account. Earlier Playground history is cleared once after the upgrade.
Claude requests get the thinking effort each model accepts. Tools and structured output sent to Gemini keep their full JSON Schema.
New Features
- The usage log page has an Export button next to Search. It downloads the current view, with the page's filters and a chosen time range, as a CSV file in the viewer's language. Users export their own logs, and administrators in the all-users view export every user's logs. Each user runs one export at a time, across tabs and, with Redis, across nodes (#7774).
- Log Maintenance adds Allow users to export logs (on by default) and Longest log time range (days) (365 by default). Regular users' log searches had no limit before. Their searches, exports, and dashboard usage and traffic charts now share this range. The page is split into Log recording, Log search and export, history log cleanup, and server log management.
- Date and time pickers, including the date range filter on log pages, work down to the second.
Fixes and Improvements
- Known Claude models get the nearest thinking effort they accept, and none when they accept none. This fixes 400s on Opus 4.5 with max or xhigh, on Sonnet 4.5 and Haiku 4.5, and when a model suffix or modifier turns thinking off. Native Claude requests without a suffix or modifier pass through as sent.
- Claude Haiku 5.5 uses adaptive thinking with medium effort by default. Converted requests drop temperature, top_p, and top_k for it. Gemini 2.5 Pro gets a 32768 thinking budget for xhigh and max.
- Tools and structured output sent to Gemini keep const, oneOf, $ref, $defs, type arrays, and non-string enums. A schema that Gemini's typed Schema can hold is sent as before, and any other schema goes in parametersJsonSchema or responseJsonSchema. Models set to Gemini API v1 get a cleaned schema and a conversion warning (#7455, #7704).
- Responses images sent to Chat Completions upstreams use the object form of image_url, so vLLM and strict gateways accept Codex screenshots and tool images (#7578).
- Chat streams converted to Claude Messages keep answer text that arrives in the same chunk as the last reasoning token (#7664). Claude Messages converted to Chat Completions keep assistant text next to tool calls (#6871).
- Claude requests forward thinking.block_binding and every anthropic-beta header value. This fixes "The block is bound to a different conversation" in multi-turn conversations whose system prompt or tools change (#7732).
- Responses WebSocket sessions send a close frame when the server ends them, so Codex sees a normal close (#7642).
- Without Redis, model rate-limit 429s carry the same error message and log line as with Redis (#7721).
- The console reloads once when a page's code fails to load, and the error page shows a status code only when there is one (#7634).
- The wallet, security, and OAuth callback pages and the user binding dialog load much less JavaScript. Linux DO bindings show the Linux DO logo.
- Date pickers and relative times follow the interface language (#7654).
- The model square shows the latest 24 hourly status samples within 7 days, with no gaps. Expression model cards show the cache read price (#7757).
- Usage log previews keep the recorded billing tiers (#7763), and task settlement logs show tiered billing (#7446).
- Pricing conversion accepts model metadata that declares endpoints as an array

Important
完善了 i18n，控制台、API 错误、邮件、通知和服务器日志都支持多语言。使用 DEFAULT_LANGUAGE 环境变量设置默认语言，不设置时为英文。
渠道新增“忽略上游证书校验”开关，可以连接使用自签证书的上游。
任务插件新增 utils.fetch，可以读取上游数据。
豆包 Seedance 1.5 Pro、2.x 和万相 3.0 支持传 -1 自动选择时长，按上游回报的时长计费。
Improved i18n. Console messages, API errors, emails, notifications, and server logs are now multilingual. Set the default language with the DEFAULT_LANGUAGE environment variable. It defaults to English.
Channels have a new Skip Upstream TLS Verification switch for upstreams that use self-signed certificates.
Task plugins can read upstream data with the new utils.fetch.
Doubao Seedance 1.5 Pro and 2.x, and Wan 3.0, accept -1 to pick the video duration automatically and bill the duration the upstream reports.
New Features
- Improved i18n. The web console shows API messages and log entries in all seven interface languages. The backend translates errors for AI clients, emails, and notifications into English, Simplified Chinese, or Traditional Chinese, using the account language first and then the request's Accept-Language header. Log entries written before this release keep their original text.
- DEFAULT_LANGUAGE (en, zh-CN, or zh-TW) sets the default language of backend messages and server logs. It defaults to en.
- Skip Upstream TLS Verification is under the HTTP settings of each channel. It covers the channel's relay requests, Responses and Realtime WebSocket, task submission and polling, plugin fetches, balance, usage, and model-list queries, and media proxied through the channel. Codex OAuth refresh and Vertex or plugin service-account tokens still verify certificates.
- Task plugins can call utils.fetch in the submit, query, and content-download hooks. It reaches the channel Base URL and the additional request domains the plugin declares, and the upload dialog lists those domains. The Task Plugin API v1 document lists the hooks and limits.
- Plugins can call utils.json.clone and export parseSubmitEventDelta with no capability declaration. Plugins that declare json-clone@1 or submit-sse-delta@1 keep loading.
- Doubao 1.3.0 and Alibaba Bailian 1.4.2 accept a duration of -1 for Seedance 1.5 Pro and 2.x and for Wan 3.0. The request reserves the model's longest duration (12 seconds for Seedance 1.5 Pro, 15 for 2.0, 30 for 2.5 and Wan 3.0) and bills the duration the upstream reports. A JSON duration on /v1/videos now reaches Ark, and Wan 3.0 smart duration works on /v1/video/generations. (#7700)
Fixes and Improvements
- Responses input converted for Claude and Gemini drops assistant text that comes after the last user message or tool result, and records a trailing_assistant_omitted warning. Codex sends such input when it retries a stream that failed after a reply finished. Claude 4.6 and later, Gemini 3.6 Flash, and Gemini 3.5 Flash-Lite and later rejected it with a 400. Chat Completions and Claude Messages requests keep their prefill.
- xAI Grok Imagine 1.1.0 bills video edits and extensions at the output resolution. The output keeps the input video's resolution up to 720p. The plugin reads the output file header and bills 480p when the shorter side is 480 pixels or less. It bills 720p otherwise, and when the header cannot be read.
- Hailuo 1.2.1 downloads MiniMax v1 videos through the file retrieve link. Those downloads returned 410 before. (#7709)
- Multipart task submissions send the Content-Type of the rebuilt body, so the boundary matches the body the upstream receives. (#7655)

What's Changed
- fix(task-plugin): set multipart content type for upstream requests by @LauZzL in https://github.com/QuantumNous/new-api/pull/7655
- feat(i18n): translate web console messages in the frontend and AI-client errors in the backend by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7712
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.42...v1.0.0-rc.43

Important
Search Billing & Task Plugins / 搜索计费与任务插件
重做了搜索函数计费。渠道可以按上游自己的方式进行网络搜索，并按上游回报的搜索次数计费。
任务插件性能大幅提升。支持插件决定是否保留 JSON 字段顺序，声明 json-order@1 即可；其他插件行为不变。
高级自定义渠道可以把 Claude Messages 和 Gemini 转到 OpenAI Responses。
新增 xAI Grok Imagine 视频任务插件。
Search-function billing has been reworked. Channels can encode Web search for their upstream and settle from the search counts the vendor reports.
Task Plugins are substantially faster. A plugin that must keep the client’s JSON field order can declare json-order@1. Other plugins are unchanged.
Advanced Custom channels can convert Claude Messages and Gemini to OpenAI Responses. An xAI Grok Imagine video Task Plugin is added.
New Features
- Reworked search-function billing, including per-Channel Web search encoding and vendor-reported search counts.
- Advanced Custom channels can convert Claude Messages and Gemini generateContent to OpenAI Responses.
- Moonshot Channels forward /v1/responses unchanged. Kimi currently serves kimi-k3 on that path.
- Added the xAI Grok Imagine video Task Plugin: grok-imagine-video, grok-imagine-video-1.5, and grok-imagine-video-1.5-lite.
- Advanced parameter overrides support Regex conditions (#7698).
- Users can be filtered by User Group.
- Task Plugins that declare json-order@1 keep the client’s JSON field order through to the upstream.
Fixes and Improvements
- Task Plugin requests are substantially faster, including passthrough submissions.
- Token counting is much faster, including long text without punctuation.
- Gemini stream usage is no longer multiplied by the number of frames (#7687).
- Responses tool-output images are sent to Gemini as images, not as text. Claude Opus 5.5 defaults to medium effort when the client sets none.
- Responses custom tools such as exec and apply_patch are kept on Claude and Gemini upstreams (#7636).
- Turning off the Thinking Suffix Adapter now also leaves effort suffixes such as -high unchanged.
- Completed image generations still bill when the Responses request later fails or is cancelled.
- Frontend build files no longer count against the page rate limit, so a cold load does not return 429 for the next session on the same IP.
- gpt-6-sol and gpt-6-luna use max_completion_tokens (#7559).
- System Settings are hidden from non-root admins (#7513).
- Fixed a blank Channels page on OriginOS 6. Fixed rounded clipping of Group names on API Keys (#7645).

What's Changed
- fix: isolate TLS configs across HTTP transports by @seefs001 in https://github.com/QuantumNous/new-api/pull/7627
- fix: use max_completion_tokens for gpt-6-sol and gpt-6-luna by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7559
- fix(web): Add class to wrap the long strings at auto-disable status tooltip content in channel manage page by @makisekuris in https://github.com/QuantumNous/new-api/pull/7604
- fix(web): hide system settings from non-root admins by @jimmyleocn in https://github.com/QuantumNous/new-api/pull/7513
- fix(i18n): keep cached zhTW interface code stable across page loads by @yenchieh in https://github.com/QuantumNous/new-api/pull/7139
- fix(web): correct translations, chart ordering, and font loading by @seefs001 in https://github.com/QuantumNous/new-api/pull/7628
- fix(relayconvert): keep Responses custom tools on Claude and Gemini upstreams by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7636
- fix(keys): prevent rounded clipping of group names by @jimmyleocn in https://github.com/QuantumNous/new-api/pull/7645
- fix(relaykit): Gemini stream token details, Responses tool-output media on Gemini, Opus 5.5 default effort by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7687
- feat(relay): web search encoding per channel and vendor-reported search billing by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7690
- feat: support regex conditions in advanced parameter overrides by @seefs001 in https://github.com/QuantumNous/new-api/pull/7698
New Contributors
- @makisekuris made their first contrib

Important
Access Tokens & Task Plugins / 访问令牌与任务插件
安全与访问里的单个面板访问令牌，改为最多 20 个具名访问令牌。创建时只显示一次，默认 30 天过期；作用域按本人已有权限勾选。升级后旧版令牌还能再用 30 天，之后会被拒绝，且不能重新生成，请在到期前换成新令牌。
管理员管理用户时，删除、改状态或角色、设置密码或权限、创建管理员、重置通行密钥、关闭两步验证、清除绑定，都需要一次性校验。额度调整，以及修改显示名、分组、备注，不需要。
任务插件改由 moejs 运行，性能大幅提升，插件接口不变。
豆包 Seedream 图片改走任务插件的 OpenAI Images（/v1/images/generations、/v1/images/edits）。新增 Seedream 5.0 Flash；各模型按自己的档位计价。任务插件收窄用量项之后，已保存的价格仍可编辑。
Security & Access replaces the single dashboard Access Token with up to 20 named Access tokens. A token is shown once at creation, expires in 30 days by default, and its Scopes stay within what the owner can already do. The Legacy token keeps working for 30 days after the upgrade, then it is rejected and cannot be regenerated. Replace it before then.
Admin user management now requires a single-use verification for delete, status or role changes, password or Permissions, creating an admin, resetting a Passkey, disabling 2FA, and clearing a binding. Quota adjustments, and edits to display name, Group, or remark, do not.
Task Plugins now run on moejs, with a large performance improvement.
Doubao Seedream images now go through the Task Plugin OpenAI Images paths (/v1/images/generations, /v1/images/edits). Seedream 5.0 Flash is added, and each model is priced on its own tiers. Prices saved before a Task Plugin narrows its usage facts stay editable.
New Features
- Replaced the single dashboard Access Token with up to 20 named Access tokens on Security & Access. Tokens are shown once, use Scopes, and expire after 30 days by default.
- Access tokens can be renamed without verification. Changing Permissions requires a single-use proof bound to that token.
- Moved Task Plugins to moejs, with a large performance improvement. The plugin API is unchanged.
- Added Seedream 5.0 Flash, and price each Seedream model on its own tiers. Seedream images are served through the Task Plugin OpenAI Images protocol.
- Highlight differing values when comparing source price expressions (#7543).
- Added a Refresh button on the Channels table.
Fixes and Improvements
- Keep Pricing Configuration editable after a usage profile drops facts. Remaining prices stay in place.
- Keep Responses custom tools when converting to Chat Completions, so tools such as exec and apply_patch are no longer dropped.
- Preserve Claude auto-mode safeguards and per-message output_config (#7598, #7561).
- Preserve the Responses WebSocket setting when saving a Channel (#7468).
- Revalidate Marketplace downloads so a republished plugin source does not fail the integrity check from a stale cache.
- Reject non-standard roles when creating a user.
- Release binaries now embed the real version instead of v0.0.0 (#7536).
What's Changed
- test(billing): fix websocket rejection refund test after pre-consume dropped output estimation by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7491
- fix(plugins/alibaba): keep upstream metadata in openai_image render by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7507
- fix(relayconvert): Responses→Chat 工具输出图片改为提升到后续 user 消息 by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7510
- fix(relayconvert): Claude tool_result 图片提升到 user 消息；Chat→Responses 流补齐 reasoning part 事件并按段切分 item by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7512
- fix(logs): keep model badge text readable and compact by @Yorick-Ryu in https://github.com/QuantumNous/new-api/pull/7504
- fix: move force operation into confirm slot so plugin-in-use dialog has one cancel button by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7528
- fix(web): keep text selection when opening a model pricing row for editing by @RedwindA in https://github.com/QuantumNous/new-api/pull/7403
- fix(ci): align ldflags -X package path with module path so release binaries embed the real version by @mslchy in https://github.com/QuantumNous/new-api/pull/7536
- fix: match current price column width to source columns so expressions wrap alike by @

Important
Task Plugins / 任务插件
任务插件源码上限提高到 8 MiB，更大的插件可以正常上传和保存。
上游提交返回 201 / 202 等成功状态时，任务会正常继续，不再被当成失败。
定价配置在任务插件收窄可选项后，仍可编辑剩下的档位和价格，不会整表打不开。
Claude / Responses 里工具结果中的图片会按图片送给上游，不再当成文本。
Task Plugin source uploads now allow up to 8 MiB, so larger plugins can be saved.
A 2xx upstream submit status such as 201 / 202 is accepted, so the task continues instead of failing.
Pricing Configuration stays editable after a Task Plugin narrows its options; remaining tiers keep their prices.
Images in Claude / Responses tool results are sent upstream as images, not as text.
New Features
- Raised the Task Plugin source upload limit to 8 MiB. After upload, Activate now is available.
- Accept any 2xx upstream status on task submission, including 201 Created and 202 Accepted (#7478).
Fixes and Improvements
- Keep Pricing Configuration editable after a Task Plugin drops enum values. Unreachable branches are skipped; remaining prices are preserved.
- Send Claude tool_result images and Responses tool-output media upstream as images, not as text (#7512, #7510).
- Keep upstream metadata in Alibaba OpenAI Images render (#7507).
- Resolve safe constants in plugin metadata previews. Isolate theme preferences in local storage. Include provider icons in Windows builds.

What's Changed
- test(billing): fix websocket rejection refund test after pre-consume dropped output estimation by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7491
- fix(plugins/alibaba): keep upstream metadata in openai_image render by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7507
- fix(relayconvert): Responses→Chat 工具输出图片改为提升到后续 user 消息 by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7510
- fix(relayconvert): Claude tool_result 图片提升到 user 消息；Chat→Responses 流补齐 reasoning part 事件并按段切分 item by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7512
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.39...v1.0.0-rc.40

Important
Task Plugins & Quota Settings / 任务插件与额度设置
任务插件现已兼容 OpenAI Images：
可承接 /v1/images/generations 和 /v1/images/edits。
请求内提交任务并等待完成后，按 OpenAI 图片格式返回。
阿里云百炼万相图片模型改走任务插件；豆包 Seedream 图片生成已接入。
New API 渠道可绑定多个上游任务插件，用于对接另一套网关里的多个任务插件。
渠道的模型重定向更清晰，应用上游模型变更前可先预览。
额度设置可配置钱包免预扣门槛和输入预扣倍率，余额足够时跳过预扣。
Task Plugins now serve OpenAI Images:
/v1/images/generations and /v1/images/edits.
The host submits the task, waits inside the request, and returns an OpenAI image response.
Unclaimed models still use the existing image relay.
Alibaba Bailian Wanxiang image models now go through Task Plugins.
Doubao Seedream image generation is added.
A New API Channel can bind multiple Upstream task plugins, for another gateway that hosts several Task Plugins.
Model Mapping is clearer, and upstream model changes can be previewed before apply.
Quota Settings add Wallet pre-consume bypass threshold, and Input pre-consume multiplier.
High enough balances skip pre-consume.
New Features
- Task Plugins can serve OpenAI Images (/v1/images/generations, /v1/images/edits).
Bailian Wanxiang image models now go through the Alibaba Task Plugin.
Doubao Seedream image generation is added.
- New API Channels can bind multiple Upstream task plugins.
- Quota Settings: Wallet pre-consume bypass threshold (USD) and Input pre-consume multiplier.
- Model management Sync metadata supports bulk field selection.
- API Keys can show and copy the Default API address / Current domain.
- Group configuration supports drag sorting and a clearer editor.
- Usage Logs show billing source and task artifacts for Task Plugin image work.
Async tasks record performance metrics when they reach a terminal status.
Fixes and Improvements
- Seedance and Wan Task Plugins price only the resolutions each model actually offers.
Unsupported resolutions are rejected at submit.
- Preview upstream model changes before applying them.
Model Mapping controls are simpler.
- Display conditional task billing prices more reliably.
Accept fractional Group ratios and tool prices.
Limit pricing inputs to four decimal places (#7453).
- Derive Response model mismatch from model names (#7464).
Restore compact cost bubbles and hide funding-source labels when Subscription is not in use (#7450, #7451).
- Forward reasoning_effort to Zhipu (#7124).
Keep mobile dialog actions on screen.
Start linked-account verification from the provider button.

What's Changed
- fix: forward reasoning_effort to zhipu upstream by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7124
- perf: avoid repeated Claude tool-name lookups by @Cylion-bot in https://github.com/QuantumNous/new-api/pull/7447
- fix: hide source labels when subscriptions are not in play by @seefs001 in https://github.com/QuantumNous/new-api/pull/7450
- refactor(log): 恢复费用气泡并用图标区分订阅来源 by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7451
- fix(web): 修复小数输入限制和日志分组选项未遮挡的问题 by @zuiho-kai in https://github.com/QuantumNous/new-api/pull/7453
- fix(log): derive response model mismatch from names stored flag by @seefs001 in https://github.com/QuantumNous/new-api/pull/7464
- feat: record perf metrics when async tasks reach terminal status by @feitianbubu in https://github.com/QuantumNous/new-api/pull/7481
New Contributors
- @Cylion-bot made their first contribution in https://github.com/QuantumNous/new-api/pull/7447
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.38...v1.0.0-rc.39

Important
Request Policies & Channels / 请求策略与渠道
系统设置新增请求策略：集中配置请求检查、会话与重试、渠道健康，并在使用日志中留下策略决策记录。
渠道创建与编辑补充快捷选项、透传说明，以及模型重定向工作台。高级自定义渠道可按路径单独开启透传。新增 vLLM、SGLang 渠道；Ollama 渠道可按渠道开启 OpenAI 兼容对话。
渠道可开启 Responses WebSocket。高级自定义、Sub2API、New API 渠道在开关打开后可用；高级自定义仅用于无转换的 /v1/responses。
System Settings now includes Request policies: Request checks, Sessions and retries, and Channel health, with policy decision records on Usage Logs.
Channel create and edit add Quick options, Pass-through guidance, and a Model Mapping workbench. Advanced Custom channels can enable Pass-through per path. vLLM and SGLang channels are added. Ollama channels can enable OpenAI-compatible chat per Channel.
Channels can enable Responses WebSocket. Advanced Custom, Sub2API, and New API channels can use it when the toggle is on. Advanced Custom only uses converter-free /v1/responses paths.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Added Request policies in System Settings: Request checks, Sessions and retries, and Channel health, with policy decision records on Usage Logs.
- Added Responses WebSocket relay, with a per-Channel toggle. Advanced Custom, Sub2API, and New API channels can use it when enabled; Advanced Custom only uses converter-free /v1/responses.
- Added a Model Mapping workbench, Quick options, and Pass-through controls in the Channel editor. Advanced Custom channels can enable Pass-through per path.
- Added vLLM and SGLang channels. Ollama channels can enable OpenAI-compatible chat per Channel.
- Added filtered task history and cleanup in System Info.
- Show subscription deductions on Usage Logs, and flag Response model mismatch when the upstream model differs.
Fixes and Improvements
- Interrupted Responses streams no longer settle at zero and refund the whole pre-consume Quota; estimated usage is billed unless upstream reported an explicit failure (#7427).
- Count Images by payload for Quota, and map image output token details into image-output billing (#7410).
- Hailuo / MiniMax Task Plugins bill only the dimensions each model family supports.
- Reserve model rate-limit slots and judge success by the response outcome, not HTTP status alone.
- Hide Task Plugin keys from the public no-Channel 503, while keeping plugin context for operators (#7414).
- Reject Gemini :countTokens as an unknown path instead of treating it as generateContent (#7388).
- Request stream usage on every cross-protocol conversion to OpenAI chat (#7389).
- Preserve Ollama tool calls from the final stream frame (#7376).
- Restore saved compatible Passkey domains; fix DeepSeek Channel balance fetch (#6814); accept mixed-case audio extensions (#7321).
- Handle legacy database constraints and scoped authorization policies on upgrade.

What's Changed
- test(web): stabilize flaky frontend suite timeouts and motion visibility races by @RedwindA in https://github.com/QuantumNous/new-api/pull/7367
- fix: unify model provider detection and add Wan icon by @seefs001 in https://github.com/QuantumNous/new-api/pull/7373
- fix(dashboard): align weekly default range with available preset by @jimmyleocn in https://github.com/QuantumNous/new-api/pull/7355
- fix(web): prevent combobox dropdowns opening on dialog autofocus by @RedwindA in https://github.com/QuantumNous/new-api/pull/7365
- fix(audio): case-insensitive file extension in GetAudioDuration by @chiliec in https://github.com/QuantumNous/new-api/pull/7321
- fix(pricing): keep model status bar spacing uniform by @jimmyleocn in http

Important
Pricing Configuration & Task Plugins / 定价配置与任务插件
旧定价模式即将弃用。定价配置可将当前价格转换为计费表达式草稿，预览生效价格后再保存。新编辑器支持按条件计价、按次、含时间条件，以及图片用量与图片缓存。gpt-image-2、gpt-image-2.5-sunburst、gpt-image-2.5-flare 已有内置计费表达式默认值；已有管理员覆盖仍然优先。
任务插件能力增强：支持流式输出；定价配置可按任务插件设置用量价格与计费表达式；插件详情展示更新日志，缺省回退英文。创建与编辑渠道时，任务插件可参与提供商与模型选择，并带出对应插件信息。
渠道的创建与编辑已统一：共用提供商选择、分页配置、模型发现、模型重定向与参数覆盖。创建渠道时会优先使用已可用的任务插件。
安全与访问支持多个通行密钥域名。通行密钥只认创建它的网站。移除仍在使用的域名会先预览受影响数量，并要求确认；被移除域名上的通行密钥将无法继续使用。
Legacy pricing will be deprecated soon. Pricing Configuration can convert current prices to an Expression draft; review the effective prices, then save to apply. The editors now support Pricing by condition, Per-request, Time-based, and image usage / image cache. gpt-image-2, gpt-image-2.5-sunburst, and gpt-image-2.5-flare have built-in Expression defaults; existing administrator overrides still take precedence.
Task Plugins capabilities are expanded: Streaming; Usage prices and Expression in Pricing Configuration per Task Plugin; and Changelog in plugin details with an English fallback. When creating or editing a Channel, Task Plugins can take part in provider and model selection and bring along the matching plugin info.
Channel create and edit are unified: shared provider selection, tabbed setup, model discovery, Model Mapping, and parameter overrides. Channel creation prefers an available Task Plugin.
Security & Access supports multiple Passkey domains. Passkeys only work on the website they were created for. Removing a domain still in use previews how many Passkeys are affected and requires confirmation; Passkeys on a removed domain will stop working.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Added Convert to expression in Pricing Configuration: review a legacy-to-Expression draft with effective prices, then save to apply it. Legacy pricing will be deprecated soon.
- Extended Expression editors and usage details: Pricing by condition, Per-request, Time-based, and image usage / image cache. Confirmation is required before replacing an Expression.
- Added built-in Expression defaults for gpt-image-2, gpt-image-2.5-sunburst, and gpt-image-2.5-flare, including separate text and image cache prices. Existing administrator overrides stay in place.
- Expanded Task Plugins capabilities: Streaming, Usage prices and Expression per Task Plugin, and localized Changelog (English fallback).
- Unified Channel create and edit: shared provider selection, tabbed setup, inline model discovery, Model Mapping, and parameter overrides.
- Let Task Plugins take part in Channel provider and model selection, including a gateway category and preferring available plugins for legacy task providers.
- Added System updates reminders for administrators.
- Added multiple Passkey domains in Security & Access, with impact preview and confirmation before removing a domain still in use.
Fixes and Improvements
- Validate image quantities before reserving Quota, and keep image usage consistent through reservation, retries, and settlement.
- Stop Pricing Configuration writes from resetting or failing to persist on some long-lived deployments.
- Allow editing multi-key selection strategy without dropping existing keys; omit the strategy for single-key Channels.
- Restore the add split button in Advanced Custom (#7289).
- Correct Wan model protocols and usage accounting; preserve provider fields in Sora video queries; fix VolcEngine Channel model-list fetch (#7203).
- Unify server error notifications so each failed operatio

Important
Task Plugins & Pricing / 任务插件与定价
本版本完善任务插件与插件市场：插件图标、插件官网、插件元数据与能力，创建渠道时可带出对应插件信息。同时改进模型管理、目录展示与广场展示、定价配置，以及 API 密钥与用户额度、使用日志、兑换码。定价货币可切换为站点货币以便录入和查看，额度仍按美元扣费。
This release improves Task Plugins and the Marketplace: Plugin icon, Plugin website, Plugin metadata, and Capabilities, plus matching plugin info when creating a Channel. It also improves Model management, Catalog visibility, Model square visibility, Pricing Configuration, Quota on API Keys and users, Usage Logs, and Redemption Codes. Pricing currency can switch to Site currency for entry and review; Quota is still deducted in USD.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Improved Task Plugins and the Marketplace: Plugin metadata, Plugin website, Plugin icon (including optional icon.svg / icon.png), integrity checks, Capabilities, and clearer install/update flows.
- Improved Channel setup for Task Plugins so a new Channel picks up the matching Plugin icon and Plugin metadata.
- Added Site currency in Pricing Configuration. Display and conversion use the site exchange rate; stored prices remain in USD.
- Improved Model management, pricing display, Catalog visibility, and Model square visibility. Channel-configured models appear without creating catalog metadata first. The Model square can show Pricing by condition, usage prices, and visibility reasons.
- Added a searchable Group filter on Usage Logs. Suggestions come from the current view; historical Group names can still be typed in.
- Refined the Usage Logs mobile layout so filters and quick actions stay available.
- Refined Quota display on API Keys and users: Remaining quota and used amounts, progress bars, Group ratio badges, and activity timestamps. Full API Keys are resolved only for explicit copy or chat actions.
- Added confirmed batch deletion for Redemption Codes, plus an optional Save as a file step (TXT / Markdown) after creation.
Fixes and Improvements
- Show desktop API Keys Remaining quota and used amounts side by side, with more room and consistent cell spacing. Mobile stacked labels are unchanged.
- Unified table cells and Quota details across list pages.

Full Changelog: v1.0.0-rc.35...v1.0.0-rc.36

Important
Task Plugin Controls / 任务插件控制简化
任务插件现在只有两层开关：总开关与每个插件自己的开关。关闭某个插件后，不会再回退到同名内置插件，同名模型也不会再继续由内置插件处理。内置插件会按请求中的模型名回显（遵循模型重定向），而不是上游真实 ID。
Task plugins now have two control layers only: the master switch and each plugin’s own toggle. Disabling a plugin no longer falls back to a same-name built-in plugin, so same-name models are no longer still handled by the built-in plugin. Built-in plugins echo the requested model name (honoring model redirects) instead of the upstream model ID.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Added Alibaba Bailian wan3.0-video and wan3.0-video-prime all-in-one video models, with DashScope media, resolution/ratio, duration rules, and billing ratios.
- Supported wan3.0 smart duration (duration / seconds = -1): submit -1 upstream, pre-consume 30 seconds, then settle from actual usage. Other Wan models reject this marker instead of silently defaulting to 5 seconds.
- Added kimi-k3 to the Moonshot model list.
Fixes and Improvements
- Preserved Kimi K3 mid-conversation tool-loading messages, so dynamic tools are no longer dropped during relay and tool_choice=required no longer fails for missing tools (#7235).
- Made wan3.0 convert and billing follow the redirected upstream model, reject unknown sizes, restore the image-only guard for text-to-video models, and forward ratio/audio in native passthrough (#7240).
- Made the SunoAPI built-in honor model redirects and fixed lyrics/music rendering. Bumped sunoapi to 1.0.2.
- Replaced the model-card health strip’s last-three-bucket rates with an hourly series, so empty hours stay gray and the 24-slot bar can show a full day.
- Portaled Combobox and Select popups into drawer content so they stay clickable inside modal drawers.
- Switched the pricing card grid (and its skeleton) to three columns at the xl breakpoint.
- Routed common and relaykit JSON helpers through a host-injectable codec so the JSON engine can be swapped in one place.

Full Changelog: v1.0.0-rc.34...v1.0.0-rc.35

Important
Account Security / 账户安全功能完善
本版本补齐账户安全能力：登录与敏感操作共用校验流程；已登记的 TOTP 与 Passkey 可作为互备因子；系统访问令牌支持查看、轮换、撤销与访问记录；新增独立审计日志。注销账户、绑定、改密、Passkey/2FA 登记与渠道密钥查看均需作用域一次性证明，并绑定到发起会话。
Telegram 登录已迁到统一 OAuth。管理员须在 BotFather Login Widget 中登记 /oauth/telegram，并填写 Client ID / Client Secret 后重新启用；已有绑定会保留。密码存储改为默认 Argon2id（兼容历史 bcrypt），是本次安全加固的一部分，而非单独的升级动作。
This release completes account security: login and sensitive operations share one verification flow; enrolled TOTP and Passkey are alternative factors; system access tokens support status, rotation, revocation, and access history; and a dedicated audit log is added. Account deletion, bindings, password changes, Passkey/2FA enrollment, and channel-key access require scoped, single-use proofs bound to the initiating session.
Telegram login now uses unified OAuth. Administrators must register /oauth/telegram in BotFather’s Login Widget and enter the Client ID / Client Secret before re-enabling it; existing bindings are preserved. Password storage now defaults to Argon2id (with bcrypt-compatible reads) as part of this hardening, not as a standalone upgrade.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Reworked model and vendor management: catalog metadata validation, vendor management, batch delete with channel/pricing cleanup, model pricing snapshot editing with optimistic concurrency, and upstream ratio sync with price cells. The models page is organized around vendors and model connections, with dedicated pricing and vendor-management dialogs.
- Moved account security settings to a dedicated page, with system access-token status, rotation, revocation, and access history. Generate and revoke require scoped, single-use verification.
- Added a dedicated audit log page (audit.read). Successful and failed API token and quota operations are recorded with role snapshots and safe metadata, correlated with top-up records. Model, vendor, and pricing mutations are also audited.
- Unified login and sensitive-operation verification: enrolled TOTP and Passkey are alternative factors. Every primary login transport is gated before a session is issued; WebAuthn requires user verification; login challenges are consumed atomically with session creation.
- Required a scoped, single-use proof for account deletion, rechecked the session inside the deletion transaction, and revoked all sessions afterward. Bindings, password changes, Passkey/2FA enrollment, and channel-key access are likewise bound to the initiating session and action.
Security
- Completed the account-security surface: shared verification UI, session-bound proofs, last usable login method preservation, and audit of security operation outcomes. OAuth callback secrets are redacted from logs.
- Applied Unicode-aware password length limits (8–128). New passwords default to Argon2id; existing bcrypt hashes remain valid. Profiles return has_password from the existing SELECT without extra queries.
- Migrated Telegram login, binding, and security verification to the unified OAuth flow (authorization code + PKCE + ID tokens). Legacy widget / bind endpoints have been removed. A custom OAuth provider must not use the reserved telegram slug.
Fixes and Improvements
- Fixed OpenAI Chat capability detection: match gpt-<n> by major version (n >= 5) so later generations such as gpt-6-astra use max_completion_tokens and drop unsupported temperature / top_p / logprobs. gpt-4.1, gpt-4o, gpt-oss, gpt-image, and gpt-realtime are unchanged (#7211).
- Upgraded the 

Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
Important
OpenRouter Compatibility / OpenRouter 兼容性说明
OpenRouter 渠道不再将任意以 -thinking 结尾的模型名自动解释为思考模式别名，以避免错误改写 kimi-k2-thinking 等真实模型 ID。
依赖旧行为的用户应改用 some-model@thinking:on，或通过渠道模型映射保留原有公开名称：{"some-model-thinking":"some-model@thinking:on"}。符合模型家族识别规则的 Claude 与 Gemini 旧式思考后缀仍受支持。
OpenRouter channels no longer interpret arbitrary model names ending in -thinking as reasoning aliases, preventing real model IDs such as kimi-k2-thinking from being rewritten.
If you relied on the old behavior, use some-model@thinking:on, or preserve the old public name with channel model mapping: {"some-model-thinking":"some-model@thinking:on"}. Recognized Claude and Gemini family thinking aliases remain supported.
New Features
- Added gpt-6-astra to the OpenAI model list with built-in expression pricing for standard and long-context requests, including input, output, cache-read, and cache-write usage. Existing administrator pricing overrides take precedence.
- Added explicit trailing @key:value model modifiers for non-pass-through relay requests:
- @thinking:on, @thinking:off, @thinking:adaptive, or @thinking:<budget>
- @effort:none|minimal|low|medium|high|xhigh|max
- @temperature:<number> and @topp:<number>
- Model modifiers override corresponding request-body fields and remain effective across channel model mapping. Modifiers on the mapped model take precedence over those on the requested model; repeated keys use the rightmost value.
- Added administrator-visible conversion diagnostics for modifier conflicts, duplicates, and provider-specific adjustments.
- Added re:-prefixed Go regular expressions to Models that skip thinking suffix processing, allowing real model names containing @key:value to bypass modifier parsing and validation. Unknown modifier keys and invalid values otherwise return HTTP 400.
Billing and Routing
- Added canonical billing names such as base@effort:high@thinking:on and base@thinking:on, with fallback to the base model when no matching variant price is configured.
- Equivalent modifier variants share canonical pricing names. Numeric thinking-budget values, temperature, and topp do not create separate canonical pricing variants.
- Updated pricing settings and pricing synchronization to expose the effective built-in billing defaults used at runtime.
- Improved channel selection, model mapping, and token model-limit matching for models with modifiers or supported legacy aliases.
- Preserved existing legacy alias pricing and Gemini thinking-budget wildcard entries.
Fixes and Improvements
- Fixed implicit reasoning-effort remapping: max and xhigh now remain distinct through OpenAI Chat and Responses conversion paths.
- Preserved native reasoning controls when no model modifier or protocol conversion requires adaptation, avoiding unnecessary normalization or rejection of provider-native values.
- Restricted generic legacy suffix parsing to recognized GPT/o-series, Claude, and Gemini model families, protecting real model names such as qwen-max and kimi-k2-thinking.
- Applied thinking-suffix exemptions consistently across DeepSeek, Volcengine, and xAI adaptors.
- Improved Claude reasoning conversion by adapting unsupported combinations where possible and recording diagnostic warnings.
- In global or channel-level pass-through mode, model modifiers are forwarded unchanged without local parsing or validation.
- Improved HTTP 503 messages for models claimed by task plugins: errors now identify the claim

Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
Important
Compatibility Notes / 兼容性说明
此前 OpenRouter 渠道会将所有以 -thinking 结尾的模型名解释为思考模式别名。该行为现已移除，以避免错误改写 kimi-k2-thinking 等真实模型 ID。
依赖旧行为的用户应改用 some-model@thinking:on，或通过渠道模型映射保留原有公开名称：{"some-model-thinking":"some-model@thinking:on"}。Claude 与 Gemini 模型家族原有的思考后缀仍然兼容。
v1.0.0-rc.31 引入的 POST /v1/messages/count_tokens 端点在本版本中暂时停用。
The previous OpenRouter-specific behavior that interpreted every model name ending in -thinking as a reasoning alias has been removed, preventing real model IDs such as kimi-k2-thinking from being rewritten.
If you relied on the old behavior, use some-model@thinking:on, or preserve the old public name with channel model mapping: {"some-model-thinking":"some-model@thinking:on"}. Existing Claude and Gemini family aliases remain supported.
The POST /v1/messages/count_tokens endpoint introduced in v1.0.0-rc.31 is temporarily disabled in this release.
New Features
- Added explicit trailing model modifiers for non-pass-through relay requests:
- @thinking:on, @thinking:off, @thinking:adaptive, or @thinking:<budget>
- @effort:none|minimal|low|medium|high|xhigh|max
- @temperature:<number> and @topp:<number>
- Model modifiers override the corresponding request-body fields and remain effective across channel model mapping. Modifiers configured on the mapped model take precedence over those on the requested model.
- Added conversion diagnostics for overridden, duplicated, or adjusted reasoning and sampling settings. These warnings are recorded with consumption logs for troubleshooting.
- Unknown modifier keys and invalid values now return HTTP 400.
- Added re:-prefixed Go regular expressions to Models that skip thinking suffix processing, allowing real model IDs containing @key:value segments to bypass modifier parsing and validation.
Billing and Routing
- Added canonical billing identities for model variants. Pricing can now target forms such as base@effort:high@thinking:on or base@thinking:on, with automatic fallback to the base model.
- Canonical billing is independent of modifier order, duplicate modifiers, and numeric thinking-budget values. temperature and topp never affect the selected billing identity.
- Updated channel selection, model listing, model mapping, and token model-limit checks to recognize the underlying base model when modifiers or supported legacy aliases are used.
- Preserved existing wildcard pricing behavior for legacy Gemini thinking-budget model names.
Relay Improvements
- Limited legacy -thinking, -nothinking, thinking-budget, and effort suffix parsing to recognized GPT/o-series, Claude, and Gemini model families, preventing ordinary model IDs such as qwen-max from being misinterpreted.
- Improved Claude reasoning conversion by safely adapting unsupported thinking modes, effort levels, budgets, and sampling combinations while recording conversion warnings.
- Native Claude and Gemini requests without host-specific model modifiers continue to preserve their provider-native reasoning controls.
- Global and channel-level pass-through requests remain byte-identical: model modifiers are not parsed or validated and are forwarded verbatim.
Bug Fixes
- Improved task-plugin HTTP 503 errors to identify the plugin that owns the requested model when it has no enabled channel in the selected group. The error now recommends disabling or overriding the claiming plugin instead of returning only a generic “no available channel” message (#7185).
- Fixed release builds resolving

Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
New Features
- Expanded request, response, and streaming conversion across OpenAI Chat Completions, OpenAI Responses, Claude Messages, and Gemini. Hosted tools, citations, grounding metadata, reasoning intent, and authoritative usage are now preserved more faithfully across supported relay paths (#7137, #7170).
- Added the Anthropic-compatible POST /v1/messages/count_tokens endpoint and enabled Claude channels to serve Gemini-format clients through protocol conversion.
- Released RelayKit v0.2.0 with conversion diagnostics and configurable tool-loss policies.
- Added MiniMax-H3 (Hailuo-03) /v2 video generation to the Hailuo task plugin, including multimodal image, video, and audio references, 768P/2K output, and validated 4–15 second durations (#7168, #7171).
- Added MiniMax-H3 usage facts for input_images and input_video_seconds. Existing ModelBillingExpr values are not migrated automatically and must reference these fields to include input-media charges.
Reliability Improvements
- Hardened asynchronous task polling with host-level HTTP classification. Upstream 404/410 responses now fail and refund immediately, while transient, authentication, hook, and unrecognized-response failures use a bounded consecutive-failure counter.
- Added TASK_POLL_MAX_FAILURES, defaulting to 20. A successful poll resets the counter; reaching the limit fails and refunds the task. Set it to 0 or lower to disable this limit. The existing TASK_TIMEOUT_MINUTES deadline remains unchanged.
- Reduced unnecessary anonymous cold-start authentication requests and added ETag/304 revalidation for public notice, home, about, and legal-content APIs (#7166).
- Fixed unchanged system-task state writes being incorrectly treated as task-lock loss (#7135).
Billing, Privacy, and Bug Fixes
- Improved billing integrity across converted and streamed requests by preserving authoritative upstream usage and fixing Claude cache accounting, Gemini modality usage, tool-call surcharges, and interrupted-stream settlement (#7137, #7170).
- Fixed MiniMax-H3 input-media accounting so bounded estimates can be settled against valid upstream image and duration usage (#7171).
- Hardened usage-log privacy by separating user, administrator, root, and audit metadata and stripping sensitive channel and rejection details from user-visible log responses.
- Corrected Video API links across localized documentation (#7116).
New Contributors
- @To3akaRin made their first contribution in #7116.
- @NebulaMao made their first contribution in #7168.
- @CreatorEdition made their first contribution in #7166.

Full Changelog: v1.0.0-rc.30...v1.0.0-rc.31

Caution
Critical Upgrade Notice / 关键升级提示
v1.0.0-rc.28 和 v1.0.0-rc.29 存在数据库迁移兼容性问题，部分使用 MySQL 或 PostgreSQL 的现有实例在升级后可能无法启动。
- v1.0.0-rc.29 无法启动或陷入重启循环：请升级至 v1.0.0-rc.30 并重启服务。
v1.0.0-rc.28 and v1.0.0-rc.29 contained a database migration compatibility regression that could prevent some existing MySQL or PostgreSQL deployments from starting after an upgrade.
- v1.0.0-rc.29 fails to start or enters a restart loop: upgrade to v1.0.0-rc.30 and restart the service.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
Bug Fixes
- Fixed usage-log statistics showing a zero quota total despite matching consumption records containing non-zero quota. This was a reporting-only issue; consumption logs, actual billing, and account balances were not affected (#7106, #7108).
- Added a safe and idempotent PostgreSQL migration for known legacy unique constraints on tokens.key, preserving existing token data and key uniqueness while preventing startup migration conflicts.
New Contributors
- @jimmyleocn made their first contribution in #7108.

Full Changelog: v1.0.0-rc.29...v1.0.0-rc.30

Caution
Critical Upgrade Notice / 关键升级提示
v1.0.0-rc.28 存在数据库迁移兼容性问题，部分使用 MySQL 或 PostgreSQL 的现有实例在升级后可能无法启动。
- 仍在 v1.0.0-rc.27：请跳过 v1.0.0-rc.28，直接升级至 v1.0.0-rc.29。
v1.0.0-rc.28 contained a database migration compatibility regression that could prevent some existing MySQL or PostgreSQL deployments from starting after an upgrade.
- Still on v1.0.0-rc.27: skip v1.0.0-rc.28 and upgrade directly to v1.0.0-rc.29.
Warning
Experimental Plugin System / 实验性插件系统
自 v1.0.0-rc.27 起引入的实验性插件系统仍处于测试阶段，并涉及较大的架构调整。如遇问题，请及时反馈。
从 v1.0.0-rc.26 或更早版本升级的用户，必须重新配置所有视频模型价格；已完成该配置的用户无需重复操作。本版本仍不推荐用于生产环境。
The experimental plugin system introduced in v1.0.0-rc.27 remains under testing and includes major architectural changes. Please report any issues promptly.
Users upgrading from v1.0.0-rc.26 or earlier must reconfigure all video model prices. Users who have already completed this step do not need to repeat it. This release is still not recommended for production use.
Bug Fixes
- Fixed database initialization failures on existing MySQL and PostgreSQL deployments caused by incompatible GORM core and database driver versions in v1.0.0-rc.28 (#7098, #7099).
- Added a safe and idempotent PostgreSQL migration for legacy prefill_groups.name uniqueness constraints, preserving existing data and the intended soft-delete behavior.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.28...v1.0.0-rc.29

⚠️ IMPORTANT WARNING / 重要警告 ⚠️
最新版已增加插件系统测试版本，系统架构大改，有 bug 请立即报告。对于切换到这个版本的用户，需要重新配置所有的视频模型价格！不推荐生产环境使用。
The latest version introduces an experimental plugin system with major architectural changes. Please report any bugs immediately. Users switching to this version MUST reconfigure all video model prices! Not recommended for production environments.
New Features
- Replaced built-in task adaptors with a new sandboxed JavaScript plugin system (#7076)
- Added support for GLM channel /v1/responses endpoint (#7050)
- Enabled passthrough of Claude Messages and OpenAI Responses for Ollama (#7051)
Bug Fixes
- Fixed an issue in billing where a constantly true time rule expression caused multipliers to apply all day (#6934)
- Restored the ability for administrators to unbind built-in providers in the web interface (#6987)
- Fixed docker development build context by adding relaykit go.mod (#7072)
Other
- Updated PR templates and removed the PR Check workflow (#7053)

What's Changed
- feat: glm chanel /v1/responses by @seefs001 in https://github.com/QuantumNous/new-api/pull/7050
- feat(ollama): passthrough Claude Messages and OpenAI Responses by @seefs001 in https://github.com/QuantumNous/new-api/pull/7051
- docs: update PR template and remove PR Check workflow by @seefs001 in https://github.com/QuantumNous/new-api/pull/7053
- fix(web): restore admin unbinding for built-in providers by @zcxads666 in https://github.com/QuantumNous/new-api/pull/6987
- fix(billing): 修复时间规则区间恒真表达式导致倍率全天生效 by @zcxads666 in https://github.com/QuantumNous/new-api/pull/6934
- fix(docker): add relaykit go.mod to dev build context by @VladKabiak in https://github.com/QuantumNous/new-api/pull/7072
- feat(task): replace built-in task adaptors with a sandboxed JS plugin system by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/7076
New Contributors
- @VladKabiak made their first contribution in https://github.com/QuantumNous/new-api/pull/7072
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.26...v1.0.0-rc.27

⚠️ Important Warnings / 重要警告
- Database Migration Risk / 数据库迁移风险：This release changes quota fields to support values larger than 4,294,967,295. Upon startup, the system will check database table structures. Startup may fail if the column types are incorrect. In such cases, manual modification is required (instructions will be in the error message; consult AI if unsure how to proceed).
本次更新修复了额度上限问题，支持大于 42 亿的数值。启动时系统会检测数据库表结构，如果不匹配可能会导致启动失败。遇到此问题需根据错误提示手动修改数据库表结构（如不熟悉数据库操作，可寻求 AI 辅助）。
- Deprecation of 32-bit Systems / 废弃 32 位系统支持：Support for 32-bit operating systems has been temporarily deprecated to accommodate these changes.
为配合上述更改，我们已“临时”废弃对 32 位操作系统架构的支持。
New Features
- Added support for thinking_token_budget in vLLM (#7027)
Bug Fixes
- Prevented browser auto-fill from incorrectly populating credentials in usage log filters (#6966)
Improvements
- Removed 32-bit integer limitations to allow quota values to exceed 4294 million (#7025)

What's Changed
- refactor(web): prevent credential autofill in usage log filters by @seefs001 in https://github.com/QuantumNous/new-api/pull/6966
- refactor: deprecate int32 by @seefs001 in https://github.com/QuantumNous/new-api/pull/7025
- feat: vllm thinking_token_budget by @seefs001 in https://github.com/QuantumNous/new-api/pull/7027
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.25...v1.0.0-rc.26

Highlights
Channel testing and gateway administration gain safer workflows, user/group-aware overrides, and request-field passthrough controls. This release also improves provider compatibility, Responses API accounting, and recharge/quota reliability.
New Features
- Channel testing has been expanded with safer test workflows, including an auto-disable-only mode for failing channels (#6917).
- Parameter overrides can now use requester user and group context for more flexible routing and policy rules (#6534).
- Gateway channels now let administrators choose which request fields are passed through to upstream providers (#6847).
Bug Fixes
- Ali provider requests now avoid sending unset top_p values and correctly handle mapped image models (#6772).
- Qwen TTS models are now classified correctly so they route through the proper text-to-speech path (#6711).
- Responses API handling now preserves penalties and prompt-cache settings while settling cached-token usage accurately (#6892).
- Ollama and Claude conversions now preserve reasoning/tool-call context while avoiding invalid empty tool payloads (#6862).
- Gemini-style model listing requests to /v1/models now return correctly (#6199).
- Recharge, refund, and quota updates are now settled more consistently, including concurrent and uncreditable order cases (#6845).
- Web forms now keep fetched model selections in sync, and the mobile sidebar behaves correctly (#6841).
- Usage logs now record reasoning effort consistently (#6641).
- Request length validation is now enforced consistently on the backend (#5548).
Improvements
- Billing logs now highlight the conditional multiplier that matched a request, making cost calculations easier to audit (#6561).
- Large server and model-list searches in the web UI are debounced, improving responsiveness on big installations (#6727).
- The playground now displays streamed output with smoother word fade-ins and a more robust editor experience (#6895).
- Custom channel route editing now has a more capable advanced editor (#6865).

What's Changed
- fix(ali): stop injecting top_p into requests that omit it by @ENCHIGO in https://github.com/QuantumNous/new-api/pull/6674
- fix(channels): classify Qwen TTS models correctly by @seefs001 in https://github.com/QuantumNous/new-api/pull/6711
- feat(channels): add auto-disable-only channel test mode by @seefs001 in https://github.com/QuantumNous/new-api/pull/6728
- perf(web): debounce server and large-list searches by @seefs001 in https://github.com/QuantumNous/new-api/pull/6727
- fix: record reasoning effort consistently in usage logs by @seefs001 in https://github.com/QuantumNous/new-api/pull/6641
- feat(relay): expose user and group context to parameter override by @seefs001 in https://github.com/QuantumNous/new-api/pull/6534
- fix(ollama): preserve reasoning and tool-call context by @seefs001 in https://github.com/QuantumNous/new-api/pull/6605
- fix: backend length validation by @seefs001 in https://github.com/QuantumNous/new-api/pull/5548
- feat(billing): highlight matched conditional multipliers in logs by @seefs001 in https://github.com/QuantumNous/new-api/pull/6561
- fix(web): require confirmation before rotating access token by @seefs001 in https://github.com/QuantumNous/new-api/pull/6749
- fix(relay): preserve presence/frequency penalty in Responses conversion by @UzkiS in https://github.com/QuantumNous/new-api/pull/6654
- refactor(responses): remove compact model suffix handling by @seefs001 in https://github.com/QuantumNous/new-api/pull/6770
- fix: mobile sidebar by @seefs001 in https://github.com/QuantumNous/new-api/pull/6760
- fix(web): refresh Turnstile token after login attempt by @thiswod in https://github.com/QuantumNous/new-api/pull/6764
- fix: 修复 Gemini 风格 /v1/models 列表请求 by @zhibisora in https://github.com/QuantumNous/new-api/pull/6199
- fix(billing): 异步任务退款时同步减少 used_quota by @wans10 in https://github.com/QuantumNous/new-api/pull/6795
- fix(web): keep fetched model selection in sync wit

Highlights: This release expands provider and channel compatibility with New API channels, DeepSeek Responses API support, zstd request decompression, and automatic group routing. It also improves operational accuracy with stream visibility in logs, more reliable tiered-retry billing, and upstream Bedrock cancellation when clients disconnect.
New Features
- Added New API channel support with Gemini-to-OpenAI streaming conversion, Advanced Custom Responses Compact handling, and multipart image edit support (#6559).
- Added DeepSeek Responses API support (#6562).
- Added automatic group selection for request routing (#6590).
- Added per-channel HTTP transport controls so administrators can tune connection behavior per channel.
- Added support for zstd-compressed requests (#6545).
- Log owners can now see whether a request was streamed (#6558).
Bug Fixes
- Qwen thinking_budget is now preserved when requests are relayed (#5836).
- Tiered retry billing now settles usage against the final selected group, including group-switch retries (#6570).
- Bedrock requests are now cancelled when clients disconnect, reducing wasted upstream work (#6589).
Improvements
- Public header navigation links now use a consistent font size (#6557).

What's Changed
- refactor: extract protocol conversion layer into standalone relaykit module by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/6369
- fix: 慢查询/错误 SQL 日志参数化 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6493
- feat(oidc): 支持自定义 OIDC 登录显示名称 by @mrchi in https://github.com/QuantumNous/new-api/pull/6012
- fix: preserve Qwen thinking_budget passthrough by @wxwwt in https://github.com/QuantumNous/new-api/pull/5836
- feat: support zstd request decompression by @seefs001 in https://github.com/QuantumNous/new-api/pull/6545
- feat(logs): expose stream status to log owners by @seefs001 in https://github.com/QuantumNous/new-api/pull/6558
- fix(relay): preserve multipart image edits for New API channels by @seefs001 in https://github.com/QuantumNous/new-api/pull/6559
- fix(oauth): stop treating a foreign window.opener as a bind flow by @neimaravila in https://github.com/QuantumNous/new-api/pull/6425
- style: unify public header nav link font size to 14px (text-sm) by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6557
- feat: deepseek responses api by @seefs001 in https://github.com/QuantumNous/new-api/pull/6562
- fix(billing): settle tiered retries with final group by @faithleysath in https://github.com/QuantumNous/new-api/pull/6518
- fix(billing): harden tiered retry group-switch billing (follow-up to #6518) by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/6570
- fix(aws): cancel Bedrock requests on client disconnect by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/6589
- Feat/auto group by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/6590
New Contributors
- @mrchi made their first contribution in https://github.com/QuantumNous/new-api/pull/6012
- @neimaravila made their first contribution in https://github.com/QuantumNous/new-api/pull/6425
- @faithleysath made their first contribution in https://github.com/QuantumNous/new-api/pull/6518
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.22...v1.0.0-rc.23

This release expands channel and model administration with upstream model discovery, new Gemini image models, configurable billing options, and Sub2API support. It also improves reliability across streaming, pricing, refunds, user management, and admin UI workflows.
New Features
- Channels can now discover available models from advanced custom and Codex-compatible upstreams (#6184).
- Added support for the latest Gemini image GA models (#6371).
- Added configurable tool pricing with Sub2API channel support and alpha-search billing.
- OpenAI Realtime GA models are now available with improved routing compatibility (#6032).
Bug Fixes
- Responses-to-Chat streaming no longer emits duplicate tool calls (#6225).
- Creating a model with an existing name no longer clears its saved pricing (#6365).
- The unset-price models view now lists only models available from the selected channels (#6126).
- Very large user quota values are now handled without overflow (#6134).
- User list sorting now works correctly with paginated results (#6194).
- MiniMax models are now correctly associated with the MiniMax vendor (#6164).
- Playground model group selectors now handle auto-group listings and long options correctly (#6163).
- Task refunds are no longer duplicated during concurrent status updates (#6074).
- Top-up amount fields can now be cleared completely while editing (#6473).
- Channel inline priority updates are more reliable (#6415).
Improvements
- Admin JSON editing now has a more consistent experience and avoids UI issues that could hide content (#6421).
- Users page search is debounced to reduce unnecessary requests and improve responsiveness (#6474).
- Proxy-backed channel requests are more compatible and stable (#6157).
- Upstream error logs now provide more useful fallback details when providers return unclear errors (#6431).

What's Changed
- fix: list only channel models in unset price models tab by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6126
- fix: purge authentication data on hard user deletion by @seefs001 in https://github.com/QuantumNous/new-api/pull/6168
- fix(users): prevent large quota values from overflowing by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6134
- fix: infer MiniMax vendor for MiniMax models by @fux-dev in https://github.com/QuantumNous/new-api/pull/6164
- fix: add server-side sorting to user list, prevent client-side sort on paged data by @DawnMoon1542 in https://github.com/QuantumNous/new-api/pull/6194
- fix(data-table): update table body immediately when toggling columns by @DawnMoon1542 in https://github.com/QuantumNous/new-api/pull/6253
- fix: prevent duplicate tool calls in Responses-to-Chat streaming by @nikuleo in https://github.com/QuantumNous/new-api/pull/6225
- feat(channel): support Codex upstream model discovery by @seefs001 in https://github.com/QuantumNous/new-api/pull/6184
- feat: support upstream model fetch for advanced custom channels by @seefs001 in https://github.com/QuantumNous/new-api/pull/5971
- refactor(auth): replace dashboard sessions with stateless tokens and session control by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/6329
- fix: 中转 gpt-realtime 系列 GA 模型时不再发送 realtime beta 标识 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6032
- fix(playground): prevent model group option stretching by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6120
- perf(data-table): optimize action column width allocation by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6135
- fix(web): show used quota for unlimited API keys by @RedwindA in https://github.com/QuantumNous/new-api/pull/6224
- fix(channel): improve proxy client compatibility and cache lifecycle by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/6157
- fix(playground): resolve auto group model listing by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6163
- fix: prevent duplicate suno task refunds via cas status update by @feitian

Highlights
The headline change is GPT-5.6 cache write billing: OpenAI cache_write_tokens are now charged at the cache-creation rate, with accounting safeguards around cached and uncached prompt tokens. This release also improves pricing workflows, Playground controls, channel table usability, and billing/logging reliability.
New Features
- Added GPT-5.6 cache write billing support, charging OpenAI cache_write_tokens at the cache-creation rate while keeping cached and uncached prompt token accounting consistent.
- Pricing pages now support group-aware dynamic calculations and avoid showing sample-only special groups (#5906).
- Model pricing settings now include an Unset Price Models tab plus status and sync filters, making it easier to find and configure models without prices (#6124).
- The Playground now includes request parameter panels for chat and generation tuning, so users can adjust model behavior directly in the UI (#6044).
- Channel lists now support manual column resizing and keep resized tables filling the available width (#5948).
- Logs now show stream timing metrics and task detail views for clearer troubleshooting.
- Added enhanced text protocol conversion and advanced custom routing capabilities for more flexible channel routing (#5825).
Bug Fixes
- Browser translation can no longer corrupt React-rendered pages (#5963).
- Custom model names that differ only by case can now be added consistently, including from multi-select controls (#6061).
- Billing now handles quota pre-consumption and image-stream disconnect edge cases more accurately.
- Subscription reset handling is more reliable under concurrent updates (#6057).
- Codex response passthrough fields now sync correctly (#6018).
- Playground model selector lists now stay synchronized after model changes.
- Custom home iframes can trigger top navigation after a user action as intended (#5955).
- Restored paste behavior for default channel connection settings.
Improvements
- Upstream price sync tables in model pricing now load and render more efficiently (#6092).
- Refined table badges, pricing layouts, sync dialogs, and default theme styling for a cleaner admin UI.
- Font assets now resolve correctly across workspace layouts.

What's Changed
- fix(web): prevent browser translation from mutating React roots by @cat0825 in https://github.com/QuantumNous/new-api/pull/5963
- feat(web): 支持渠道列表手动调整列宽 by @zuiho-kai in https://github.com/QuantumNous/new-api/pull/5948
- fix(web): let channel table fill available width by @zuiho-kai in https://github.com/QuantumNous/new-api/pull/6031
- chore: remove dead files resurrected by v1.0 launch commit by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6041
- fix: use GORM v2 row locking for subscription resets by @seefs001 in https://github.com/QuantumNous/new-api/pull/6057
- feat(playground): add playground parameter settings panel by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6044
- fix: allow adding model names that differ only by case in multi-select by @feitianbubu in https://github.com/QuantumNous/new-api/pull/6061
- fix: remove sample special usable groups leaking into pricing page by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5906
- perf(model-pricing): optimize upstream price sync tables by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/6092
- fix(web): allow user-activated top navigation for custom home iframe by @Heykode in https://github.com/QuantumNous/new-api/pull/5955
- chore(deps): bump golang.org/x/crypto from 0.51.0 to 0.52.0 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/6096
- fix: sync codex responses passthrough fields by @seefs001 in https://github.com/QuantumNous/new-api/pull/6018
- feat: enhance text protocol conversion and advanced custom routing by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5825
- feat: add unset price models tab to model pricing settings by @feitianbubu in https://github.com/QuantumNous/new-ap

Highlights
This release adds GPT-5.6 pricing support and stale instance cleanup tools for administrators. It also fixes several quota and pricing edge cases, including wallet reward transfers and decimal ratio input.
New Features
- Added token pricing support for GPT-5.6 models.
- Added system-info actions for cleaning up stale instances. (#5953)
Bug Fixes
- Wallet reward transfers now respect the configured quota unit instead of using an incorrect fixed unit. (#5808)
- Group ratio fields now allow decimal values such as 0.05 to be typed normally. (#5995)
- Tiered pricing now uses a default token estimate during pre-consumption, improving quota calculation for affected pricing rules.

What's Changed
- fix: 分组倍率 Ratio 输入框保留小数中间态，修复无法输入 0.05 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5995
- feat(system-info): add stale instance cleanup actions by @seefs001 in https://github.com/QuantumNous/new-api/pull/5953
- fix(wallet): honor configured quota units for reward transfers by @A-Words in https://github.com/QuantumNous/new-api/pull/5808
New Contributors
- @A-Words made their first contribution in https://github.com/QuantumNous/new-api/pull/5808
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.19-i18nfix.2...v1.0.0-rc.20

Fix i18n error

Highlights
This release improves channel testing, redemption management, theme handling, and several web UI flows. It also fixes usage reporting, classic web builds, streaming disconnect handling, and quota/video duration calculations.
New Features
- Redemption management now supports status filtering and cleanup actions for easier administration.
- Group ratio editing is easier to use, with clearer visibility behavior and more robust JSON input handling.
- Redeem failures now show clearer messages so users can understand what went wrong.
Bug Fixes
- AI Elements integrations now read nested usage data so token usage is reported correctly.
- Classic web builds now complete successfully again (#5892).
- Field passthrough controls now appear correctly for Codex channels (#5902).
- Home iframes now stay in sync with the selected theme and language (#5917).
- Custom HTML rendered inside Shadow DOM now follows light and dark mode changes correctly (#5890).
- Logged-in users are now redirected away from the sign-up page consistently with the login and classic themes (#5910).
- Streaming responses no longer continue writing after the client disconnects (#5710).
- Quota settlement and Ali video duration handling were corrected after task balance settlement (#5923).
- Build date generation no longer fails when DNS lookup fails (#5945).
Improvements
- Channel and model test dialogs are more compact and easier to operate (#5886).
- Dynamic pricing details now use styling consistent with the log details dialog (#5905).
- Mobile user cards have been refined for better readability and layout.
- Legacy top-up warning text is clearer and easier to understand (#5855).

What's Changed
- chore(deps-dev): bump tmp from 0.2.5 to 0.2.7 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/5147
- chore(web): update default web dependencies by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5884
- perf(web): refine channel test dialog interactions by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5886
- fix(classic): fix classic web build failures by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5892
- fix(channels): show field passthrough controls for Codex by @seefs001 in https://github.com/QuantumNous/new-api/pull/5902
- fix: align dynamic pricing style with log details dialog sections by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5905
- fix: 任务差额结算后 quota 和阿里视频时长优化 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5923
- fix(web): sync home iframe theme and language by @guoruqiang in https://github.com/QuantumNous/new-api/pull/5917
- fix(web): 修复 Shadow DOM 隔离渲染下自定义 HTML 深浅色失效的问题 by @olwater in https://github.com/QuantumNous/new-api/pull/5890
- fix(web): 注册页补充已登录重定向守卫，与登录页 / classic 主题保持一致 (#5908) by @Heykode in https://github.com/QuantumNous/new-api/pull/5910
- feat: optimize legacy top-up warning banner copy (#5851) by @Gravirei in https://github.com/QuantumNous/new-api/pull/5855
- fix: avoid stale stream writes after client disconnect by @seefs001 in https://github.com/QuantumNous/new-api/pull/5710
- Fix/build date dns error by @seefs001 in https://github.com/QuantumNous/new-api/pull/5945
New Contributors
- @Heykode made their first contribution in https://github.com/QuantumNous/new-api/pull/5910
- @Gravirei made their first contribution in https://github.com/QuantumNous/new-api/pull/5855
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.16...v1.0.0-rc.17

Highlights
This release expands model and media compatibility with Responses/Chat interoperability, Doubao Seedance 2.0 updates, and Wan2.7 i2v media mapping. It also improves the web experience around Playground rendering, channel management, page state retention, and graceful restarts.
New Features
- Added Responses API to Chat compatibility so providers and clients using either format can interoperate more reliably (#5787).
- Expanded Doubao Seedance 2.0 support with material asset APIs, resolution-aware billing, video-input billing, 4K billing, and new request options (#5824).
- Added Wan2.7 image-to-video media mapping support (#4984).
- Added graceful shutdown support to reduce interrupted replies and avoid losing cached panel data during restarts (#4258).
- Improved channel management with a clearer layout and a more capable advanced custom route editor.
Bug Fixes
- Fixed the model pricing tiered expression editor so it refreshes correctly when switching between models (#5752).
- Fixed async task usage logs so usage is attributed to the node that initiated the task (#5684).
- Fixed theme switching so the app returns to the home page instead of landing on a 404 route (#5612).
- Fixed custom HTML and Markdown rendering so content keeps consistent styling and spacing, including isolated HTML views (#5795).
- Fixed page state resets when switching tabs within the same route, including dashboard time ranges (#5796).
- Fixed channel status filters so they persist across page navigation (#5863).
- Fixed Ollama non-streaming tool call handling (#5865).
- Fixed Waffo product information and webhook behavior (#5704).
Improvements
- Improved the Playground chat experience and Markdown rendering (#5217).
- Streamlined table actions and destructive confirmation dialogs for a cleaner web UI (#5645).
- Improved ClickHouse log filtering for LIKE queries.
- Optimized the frontend build pipeline for faster web builds (#5786).

What's Changed
- chore(deps): bump dompurify from 3.4.5 to 3.4.11 in /web/default by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/5718
- fix(ci): install classic workspace dependencies for releases by @seefs001 in https://github.com/QuantumNous/new-api/pull/5719
- perf(web): streamline table actions and destructive dialogs by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5645
- fix: add Waffo goods info and webhook SDK update by @zhongyuanzhao-alt in https://github.com/QuantumNous/new-api/pull/5704
- fix(model-pricing): refresh tiered expression editor when switching models by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5752
- fix: sync bun.lock with dompurify 3.4.11 to restore frozen-lockfile build by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5738
- fix(theme): 切换前端主题后重置到首页，避免路由 404 by @zhangzhichaolove in https://github.com/QuantumNous/new-api/pull/5612
- fix(task): attribute async task usage log to the initiating node by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5684
- feat: add Casbin admin permissions by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5755
- feat(playground): improve Playground chat experience and Markdown rendering by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5217
- fix(auth): align password validation copy by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5759
- fix(web): render custom HTML and Markdown content consistently by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5760
- fix(openai): harden Chat-to-Responses compatibility by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5772
- feat: support Responses to Chat by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5787
- feat: doubao-seedance-2-0-260128 支持按分辨率区分计费 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5300
- feat: support doubao seedance 2.0 safety_identifier/priority and 4k billing by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5824
- fi

Highlights
This release adds several operations-focused controls, including passive channel monitoring, ClickHouse-backed log storage, system task execution, and instance visibility. It also improves frontend reliability and polish across usage logs, markdown rendering, channel selection, and optional wallet UI behavior.
New Features
- Added passive channel monitoring mode for checking channel health without actively disrupting traffic (#5592).
- Added support for storing logs in ClickHouse for deployments that need scalable log storage and querying (#5663).
- Added a system task runner with persistent cleanup progress tracking for long-running maintenance jobs (#5680).
- Added a system instance information panel to make deployment/node status easier to inspect (#5716).
- Added a user token limit configuration section in system settings (#5678).
- Added channel testing controls for environment selection and asynchronous polling delay configuration.
- Default node names now fall back to the machine hostname when NODE_NAME is not set (#5659).
Bug Fixes
- Usage logs now show provider icons correctly for CN models (#5631).
- Corrected the Stepfun provider display name (#5636).
- Fixed SMTP compatibility with STARTTLS mode and NTLM authentication (#5426).
- Fixed classic theme build failures related to timezone date handling (#5676).
- Improved markdown rendering and expanded syntax support in the web UI (#5689).
- Fixed channel card selection state so the UI stays in sync with the selected channel (#5700).
- Hid the wallet entry from the profile dropdown when the wallet module is disabled (#5708).
Improvements
- Simplified log cleanup behavior as part of the new ClickHouse logging and system task maintenance flow.

What's Changed
- fix(usage-logs): show provider icons for CN models (align with classic frontend) by @DeliciousBuding in https://github.com/QuantumNous/new-api/pull/5631
- fix: StepFun → Stepfun by @seefs001 in https://github.com/QuantumNous/new-api/pull/5636
- feat: add passive channel monitoring mode by @exherb in https://github.com/QuantumNous/new-api/pull/5592
- feat: default node name to hostname when NODE_NAME unset by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5659
- feat: support ClickHouse log database by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5663
- fix: support SMTP STARTTLS mode and NTLM auth by @bensonfx in https://github.com/QuantumNous/new-api/pull/5426
- chore(deps): bump github.com/ClickHouse/ch-go from 0.58.2 to 0.65.0 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/5664
- fix: date-fns-tz classic theme build error by @seefs001 in https://github.com/QuantumNous/new-api/pull/5676
- feat: add system task runner by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5680
- feat: add system instance info panel by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/5716
- fix(web): replace default markdown renderer and expand syntax support by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5689
- fix(web): sync channel card selection state by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/5700
- fix(web): hide wallet entry in profile dropdown when wallet module disabled by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5708
- feat(system-settings): add user token limit configuration section by @RedwindA in https://github.com/QuantumNous/new-api/pull/5678
New Contributors
- @DeliciousBuding made their first contribution in https://github.com/QuantumNous/new-api/pull/5631
- @exherb made their first contribution in https://github.com/QuantumNous/new-api/pull/5592
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.14...v1.0.0-rc.15

Highlights
This release adds a new dashboard traffic flow Sankey chart for visualizing token routing and flow patterns, with interactive highlighting, persistent filters, and limits to keep large graphs readable. Channel management also gets a faster, cleaner card layout for easier browsing and reduced UI re-rendering.
New Features
- Added a dashboard traffic flow Sankey chart to help admins understand token flow and routing behavior at a glance (#5465).
- Added routing reliability management for better control and visibility over routing behavior.
Gallery
Traffic flow chart Channel management card layout

Bug Fixes
- Fixed log filtering by login type in the new frontend (#5620).
- Fixed the billing history dialog layout so pagination no longer overlaps the list (#5627).
- Fixed model pricing entries that could not be deleted from the visual pricing list (#5628).
- Fixed repeated database column migrations running again after restarts (#5361, #5616).
Improvements
- Improved the channel management card layout and reduced unnecessary re-renders for a smoother experience.
- Simplified the model list for easier scanning and management.
- Updated channel icons for better visual consistency.

What's Changed
- fix: stop repeated allow_balance_pay column migration on restart by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5361
- fix: stop repeated allow_wallet_overflow column migration on restart by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5616
- feat(dashboard): 增加分流图，查看 token 流量。add traffic flow sankey chart by @Quaternijkon in https://github.com/QuantumNous/new-api/pull/5465
- fix(log): 修复新前端日志按登录类型(type=7)筛选失效 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5620
- fix(wallet): 修复计费历史弹窗分页与列表重叠 (#5624) by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5627
- fix(model-pricing): 修复模型定价列表项无法删除 (#5623) by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5628
New Contributors
- @Quaternijkon made their first contribution in https://github.com/QuantumNous/new-api/pull/5465
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.13...v1.0.0-rc.14

v1.0.0-rc.13 focuses on a more polished and responsive frontend experience, especially around channel management and data-heavy pages. The channel page now supports a card-style display with improved mobile layouts, while logs, metrics, endpoint selection, date picking, and usage views received UI/UX refinements.
New Features
- Added a reusable card/table view toggle for data tables, including a new card layout that makes channel and table-heavy pages easier to browse on different screen sizes.
- Added locale-aware number formatting so numeric values are displayed more naturally for users in different regions.
Improvements
- Improved the channel page card layout with a clearer left/right structure and better mobile handling for small screens.
- Refined logs, metrics, endpoint selection, date picker, toolbar controls, and model drawer interfaces for a cleaner and more consistent UI/UX.
- Updated the Codex usage interface for better readability and usability (#5595).
- Reduced unnecessary data refreshes when returning focus to the browser window, improving frontend performance.
- Updated frontend branding copy and visual styling for a more consistent product experience.
Bug Fixes
- Improved timestamp precision in drawing-related displays by including milliseconds where needed.

What's Changed
- chore(web): enforce default frontend oxc quality checks by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5585
- refactor: codex usage ui by @seefs001 in https://github.com/QuantumNous/new-api/pull/5595
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.12...v1.0.0-rc.13

Highlights: This release introduces AdvancedCustom channels for more flexible custom routing, makes channel batch testing easier to act on, and expands model marketplace performance monitoring with clearer success-rate and latency insights.
New Features
- Added AdvancedCustom channels with improved custom route matching for more flexible provider and model routing (#5590).
- Enhanced channel model batch testing with complete test execution, refreshed status updates, clearer results, failed-model deletion, and copying selected models (#5532).
- Expanded model marketplace performance monitoring with uptime calculations, recent success rates, success-rate grading, and compact latency/throughput displays.
- Added persistent table preferences so page size and column visibility are remembered across navigation and sessions (#5403).
Bug Fixes
- Fixed tool-call handling so empty arguments values still preserve the expected tool_use block (#5543).
- Fixed the CC Switch model selector so model choices display and update correctly (#5515).
- Fixed several UI clipping and overflow issues in lists, data tables, usage log details, and custom home iframes (#5510, #5536, #5533, #5545).
Improvements
- Improved payment and recharge UI with clearer payment method buttons, icons, and minimum top-up information.
- Improved subscription plan presentation with purchase descriptions, wallet overflow handling, and downgrade group support.
- Refined settings, overview setup guide, wallet/profile interactions, and log statistic card layouts for better readability.
Gallery
Advanced Custom Channel Better model batch testing

What's Changed
- fix(web): prevent list cell text and badge overflow by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5510
- fix(keys): refresh API key form options by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5512
- fix: fix CC Switch model selector by @seefs001 in https://github.com/QuantumNous/new-api/pull/5515
- fix(data-table): capture row selection state in memo comparison by @RedwindA in https://github.com/QuantumNous/new-api/pull/5524
- fix(channels): refresh channel test dialog status by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/5517
- fix: restore custom home iframe viewport height by @seefs001 in https://github.com/QuantumNous/new-api/pull/5545
- fix(web): prevent usage log details dialog from clipping bottom content by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/5533
- fix: aws ak/sk context management by @seefs001 in https://github.com/QuantumNous/new-api/pull/5547
- test: clean up reward-hacking backend tests by @seefs001 in https://github.com/QuantumNous/new-api/pull/5563
- fix: 工具调用 arguments 为空时保留 tool_use 块 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5543
- fix: complete channel model batch testing by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/5532
- feat: 持久化表格每页条数，跨导航/会话记忆 (#5377) by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5403
- fix(data-table): prevent component cell clipping by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5536
- fix(oauth): clear bindings when hard deleting users by @seefs001 in https://github.com/QuantumNous/new-api/pull/5582
- feat: advanced custom channel by @seefs001 in https://github.com/QuantumNous/new-api/pull/5590
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.11...v1.0.0-rc.12

Highlights: This release adds support for Claude Opus 4.8, improves OpenAI image API compatibility, and makes the model pricing editor and data-heavy tables easier to use. It also fixes several UI, quota display, relay compatibility, and provider-specific issues.
New Features
- Added support for Claude Opus 4.8 (#5177).
- Added an option to control whether channel stickiness is cleared when a channel is disabled (#5306).
- Added a subscription balance redemption toggle for more flexible balance handling.
- Added user ID visibility on the profile page (#5317).
- Added support for streaming image relay and image edits in the OpenAI images API (#4608).
Bug Fixes
- Forms now focus the first validation error after submission, making errors easier to find and correct (#5163).
- Fixed custom cc-switch names being reset after losing focus (#5170).
- Fixed custom model icons not staying in sync in pricing views (#5224).
- Fixed advanced channel validation errors so they are shown without incorrectly expanding unrelated settings (#5239).
- Fixed quota and balance displays so currency conversion updates both the unit and the displayed value (#5296).
- Fixed Dify remote-image handling in file-based requests (#5134).
- Fixed model resolution for video generation task lookups (#5133).
- Improved Anthropic-compatible GLM relay behavior to avoid compatibility failures (#5307).
- Narrowed OpenAI o-series model adaptation to avoid applying it to unsupported models (#5293).
- Fixed multiselect popovers so they respect the active theme (#5328).
- Fixed Kimi K2.6 temperature normalization (#5390).
- Added support for filtering deleted users by status (#5464).
- Fixed deployment settings loading so they are only fetched on the deployments page (#5466).
Improvements
- Improved admin log and batch update performance for smoother large-scale operations (#5116).
- Improved the model pricing editor with clearer layouts, better draft saving, reusable JSON editing, and support for six-decimal pricing steps (#5275).
- Improved channel search responsiveness with debouncing and better IME input handling (#5393).
- Improved frontend table rendering, pinned columns, table layout, and badge display (#5405).
- Improved dialog sizing and footer layout across the frontend (#5346).
- Simplified public page hero copy for a cleaner landing page (#5339).

What's Changed
- fix(web): improve form validation error focus by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5163
- feat(keys/cc-switch-dialog): 修复自定义 cc-switch 名称失焦后重置问题 by @GnezIew in https://github.com/QuantumNous/new-api/pull/5170
- fix(model): correct idx_created_at_id index column order to (created_at, id) by @realRoc in https://github.com/QuantumNous/new-api/pull/5191
- perf: order admin logs by created_at to use composite index by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5116
- chore(deps): bump axios from 1.15.2 to 1.16.0 in /web/classic by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/5185
- feat: support claude-opus-4-8 by @hjxwz123 in https://github.com/QuantumNous/new-api/pull/5177
- fix(pricing): sync custom model icons by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5224
- chore(web): centralize shared frontend dependency versions by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5227
- feat(web): support classic Rsbuild dev and build by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5232
- fix(channels): reveal advanced validation errors by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/5239
- docs: refine issue templates by @seefs001 in https://github.com/QuantumNous/new-api/pull/5271
- fix: 修复余额显示时只切换了单位未切换数值 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/5296
- fix(dify): initialize file pointer before remote-image field assignment by @Pluviobyte in https://github.com/QuantumNous/new-api/pull/5134
- fix(distributor): resolve model for GET /v1/video/generations/:task_id by @Pluvioby

New Features
- Added support for request_header key source (#4903).
Bug Fixes
- Batch fixed multiple UI issues in the default web interface (#4880, #4893, #4817, #4877, #4898, #4876).
- Addressed an issue where channel remarks could not be cleared (#4886).
- Corrected the filtering behavior for usage logs (#4883).
- Fixed validation for unlimited API key quotas in the web interface (#4881).
- Enforced access control on header navigation for public modules (#4889).
- Applied group filters correctly to channel list queries (#4847, #4885).
- Fixed an ambiguity issue with column names during PerfMetric upserts (#4684).
- Added missing analytics placeholders to the default theme (#4928).
- Fixed forgot password form submission by properly typing the button (#4910).
- Normalized the layout of model detail tabs in the web interface (#4938).
- Fixed user creation and deletion handling in the management panel (#4818).
- Gated the sign-up link on the login page based on registration settings (#4871).
- Updated pagination button labels in the ModelCardGrid (#4675).
- Localized the reset password confirmation message (#4769).
- Made recharge preset selections visible in dark mode (#4897).
- Fixed an issue where the API info color dot showed the wrong color due to a semantic token mismatch (#4824).
- Corrected silent failures when saving API info with invalid colors (#4823).
- Used the correct aff_code field name in registration payloads (#4965).
- Made the faceted filter popover width adaptive in data tables (#4966).
- Resolved stale form linkage issues when submitting keys (#4967).
Improvements
- Improved performance by avoiding eager formatting in debug log calls (#4929).

What's Changed
- fix: allow clearing channel remark by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4886
- fix: correct usage logs filtering by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4883
- fix(web): handle unlimited API key quota validation by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4881
- fix: enforce header nav access control for public modules by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4889
- fix: GetAllChannels ignores group filter parameter by @draftgo in https://github.com/QuantumNous/new-api/pull/4847
- fix: qualify column names in PerfMetric upsert to avoid PG ambiguity by @SAY-5 in https://github.com/QuantumNous/new-api/pull/4684
- fix: apply group filter to channel list queries by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4885
- feat: support request_header key source by @seefs001 in https://github.com/QuantumNous/new-api/pull/4903
- fix: 修复新 UI 语言与文案显示问题 by @QuentinHsu in https://github.com/QuantumNous/new-api/pull/4876
- fix: add umami & ga4 placeholder in index.html of default theme by @NitroRCr in https://github.com/QuantumNous/new-api/pull/4928
- perf: avoid eager formatting in debug log calls by @seefs001 in https://github.com/QuantumNous/new-api/pull/4929
- fix: add type="submit" to forgot password form button by @lihan3238 in https://github.com/QuantumNous/new-api/pull/4910
- fix(web): normalize model detail tabs layout by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4938
- fix(user): fix user create and delete handling in management panel by @bugmaker2 in https://github.com/QuantumNous/new-api/pull/4818
- fix(auth): expose register_enabled in /api/status and gate sign-up link on login page by @Gnonymous in https://github.com/QuantumNous/new-api/pull/4871
- fix(web/default): clarify pagination button labels in ModelCardGrid by @ellisfan in https://github.com/QuantumNous/new-api/pull/4675
- fix(auth): localize reset password confirmation by @neow021 in https://github.com/QuantumNous/new-api/pull/4769
- fix(web/default/wallet): make recharge preset selection visible in dark mode by @xinnyu in https://github.com/QuantumNous/new-api/pull/4897
- fix(web/default): api-info color dot shows wrong color due to semantic token mismatch by @Micah-Zheng in http

Highlights
This release improves observability by recording upstream request IDs in logs and adds a compliance confirmation step for paid features. If your logs table is large, pre-apply the upstream_request_id column and index with online DDL before upgrading to avoid a potentially long migration.
New Features
- Added upstream request ID tracking in request logs for easier upstream call tracing, while avoiding unintended response header overrides.
- Added a compliance confirmation step before paid features can be used.
Bug Fixes
- Fixed the channel combobox so focusing it no longer hides valid channel options (#4829).
Other
- Upgrade notice: the logs table now includes a new upstream_request_id column and index; on large installations, run the appropriate online DDL below before upgrading to avoid a long or blocking automatic migration.
MySQL / InnoDB:
-- 1. Add the column using online DDL.
ALTER TABLE logs
ADD COLUMN upstream_request_id VARCHAR(128) NOT NULL DEFAULT '',
ALGORITHM=INPLACE, LOCK=NONE;

-- 2. Add the index using online DDL.
ALTER TABLE logs
ADD INDEX idx_logs_upstream_request_id (upstream_request_id),
ALGORITHM=INPLACE, LOCK=NONE;
PostgreSQL 11+:
-- 1. Add the column without rewriting the table on PG 11+.
ALTER TABLE logs
ADD COLUMN IF NOT EXISTS upstream_request_id VARCHAR(128) NOT NULL DEFAULT '';

-- 2. Add the index concurrently to avoid blocking reads and writes.
CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_logs_upstream_request_id
ON logs (upstream_request_id);

What's Changed
- chore(deps): bump axios from 1.15.0 to 1.15.2 in /web/classic by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4634
- chore(deps-dev): bump ip-address from 10.1.0 to 10.2.0 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4811
- 修复渠道列表显示不全问题 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4829
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.5...v1.0.0-rc.6

Highlights
The default dashboard has been refreshed with built-in model performance metrics and a clearer health view. This release also adds DeepChat deeplink support and fixes several default UI, wallet, playground, and ratio-display regressions.
New Features
- Dashboard overview has been redesigned and now includes model performance metrics with a clearer performance health panel.
- DeepChat deeplinks are now supported, making it easier to open supported chat links directly in DeepChat. (#4668)
Bug Fixes
- Top-up gateway availability now uses the correct top-up configuration, so wallet payment options are shown accurately. (#4599)
- Group ratio displays now honor configured group-to-group ratio settings. (#4772)
- Default UI dropdown menu items now trigger their selection handlers reliably. (#4787)
- The playground now handles legacy classic-format messages without failing. (#4650)
- Default frontend pages now avoid several v1 UI regressions, including stray empty description areas and layout issues. (#4802)
Improvements
- The default web UI and charts now follow shared theme presets for more consistent colors, controls, and corner radii.
- Docker images now include project and third-party license notices for easier compliance review. (#4739)

What's Changed
- feat: add DeepChat deeplink support by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4668
- fix: exclude THIRD-PARTY-LICENSES.md from .dockerignore for Docker build by @yeying-xingchen in https://github.com/QuantumNous/new-api/pull/4739
- fix(wallet): read topup gateway flags from topupInfo instead of status by @ImogeneOctaviap794 in https://github.com/QuantumNous/new-api/pull/4599
- fix(web/default): use getUserGroups for ratio display to respect GroupGroupRatio by @Micah-Zheng in https://github.com/QuantumNous/new-api/pull/4772
- fix(default): support DropdownMenuItem onSelect by @lihan3238 in https://github.com/QuantumNous/new-api/pull/4787
- fix(web/default): guard playground messages against legacy classic shape by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4650
- fix: defaut ui triage by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/4802
New Contributors
- @yeying-xingchen made their first contribution in https://github.com/QuantumNous/new-api/pull/4739
- @lihan3238 made their first contribution in https://github.com/QuantumNous/new-api/pull/4787
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.4...v1.0.0-rc.5

Highlights
OpenAI image editing is more reliable: image edit requests now preserve reference image data and related metadata, preventing edits from losing their intended context (#4646). This release also adds model performance badges and fixes several billing and dropdown UI issues.
New Features
- Added model performance badges to model lists so users can compare available models at a glance.
Bug Fixes
- Fixed OpenAI image edit requests so reference images and related edit metadata are preserved when forwarded (#4646).
- Fixed billing settings forms so top-up and billing configuration inputs behave more reliably.
- Fixed dropdown and select controls so options render and select correctly after the UI component update (#4655).
Breaking Changes
- Integrations should now read the top-up link from the top-up information API instead of the status API.
Gallery
Model availability rankings Refreshed pricing and settings UI

What's Changed
- fix: preserve OpenAI image edit reference fields by @seefs001 in https://github.com/QuantumNous/new-api/pull/4646
- fix: migrate select to Base UI items API by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4655
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.3...v1.0.0-rc.4

Highlights
v1.0.0-rc.3 adds a new model rankings experience with model availability and performance visibility, making it easier to compare reliability before choosing routes or providers. This release also refreshes key default UI areas, including pricing/settings pages and table controls.
New Features
- Added a model rankings dashboard with real availability and performance data so users can compare model reliability at a glance (#4633).
Improvements
- Refreshed the default web UI with Base UI components, updated theme presets, and cleaner table toolbar interactions (#4633).
- Reorganized the system settings pricing interface to make pricing configuration easier to scan and manage.
Bug Fixes
- Fixed channel table sorting so server-side sorting now works correctly (#4600).
- Corrected subscription payment information display in the default UI.
Gallery
Model availability rankings Refreshed pricing and settings UI
Screenshot placeholder Screenshot placeholder

What's Changed
- feat(web/default): unified UI overhaul — Base UI migration, theme presets, rankings dashboard, and table toolbar refactor by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/4633
- feat: collect model performance metrics by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/4635
- fix: enable channel table server-side sorting by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4600
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.2...v1.0.0-rc.3

Highlights
v1.0.0-rc.2 officially opens up switching between the new and classic frontend, making it easier to test the redesigned UI, compare workflows, and fall back instantly when needed. This RC also refreshes key console pages and fixes several web issues affecting email binding, tiered pricing, API key grouping, and custom Vertex base URLs.
New Features
- Added a built-in switch between the new and classic frontend, so teams can adopt the redesigned UI gradually without losing access to the previous workflow.
Improvements
- Refreshed the web console with redesigned pricing pages, more consistent dashboard and log controls, better table and analytics filtering, smoother navigation, and improved mobile responsiveness.
Bug Fixes
- Fixed email binding in the redesigned frontend so the action completes correctly (#4551).
- Fixed tiered pricing number input behavior so editing values is more reliable (#4536).
- Fixed tiered pricing display issues that could cause values and labels to appear incorrectly (#4530).
- Fixed API key grouping in the default web theme so keys appear under the expected groups (#4548).
- Fixed custom Vertex base URLs so requests route correctly when a gateway prefix is configured (#4200).
Gallery
Refreshed pricing page Dashboard and analytics controls

What's Changed
- fix: 修复默认主题 API 密钥分组行为 by @Micah-Zheng in https://github.com/QuantumNous/new-api/pull/4548
- bug fix: 新版前端 email 绑定错误 by @SHLE1 in https://github.com/QuantumNous/new-api/pull/4551
- fix: improve tiered pricing number input editing by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4536
- fix(vertex): honor custom base_url as gateway prefix by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4200
- fix(web): 修复阶梯计费 Base64 解码失败与标签不匹配导致的显示错误 by @wans10 in https://github.com/QuantumNous/new-api/pull/4530
New Contributors
- @SHLE1 made their first contribution in https://github.com/QuantumNous/new-api/pull/4551
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-rc.1...v1.0.0-rc.2

v1.0.0-rc.1 is the first release candidate for the upcoming 1.0 release, and its main focus is the brand-new v1 frontend rebuilt from the ground up for a cleaner, faster, and more capable admin experience.
This RC refreshes the interface across navigation, configuration, analytics, and logs, making day-to-day management much easier as the project moves toward the final v1 launch.
New Features
- Introduced the rebuilt v1 frontend with modern navigation and layouts, a visual channel editor for parameter overrides, and easier access to system settings and theme switching (#4265).
Improvements
- Upgraded dashboards and charts with clearer model analytics, better data sorting, and more informative tooltips for easier usage analysis.
- Improved usage logs with log type indicators, usernames and avatars, and cleaner tables that make activity easier to scan.
- Polished shared layouts, tables, and form styling across the default frontend for a cleaner and more consistent experience.
Bug Fixes
- Fixed themes loaded from the database so the selected frontend theme now applies correctly at runtime (#4518).
- Fixed forwarded requests so empty reasoning fields are preserved instead of being silently dropped (#4520).
- Fixed required-field markers so forms more consistently show which inputs are mandatory (#4533).

What's Changed
- 🚀 feat: launch v1.0 — next-generation frontend built from the ground up by @t0ng7u in https://github.com/QuantumNous/new-api/pull/4265
- fix: 修复 Message.ReasoningContent/Reasoning 空思考内容在请求转发时被静默丢弃 by @heimoshuiyu in https://github.com/QuantumNous/new-api/pull/4520
- fix: sync theme to runtime when loaded from database by @Micah-Zheng in https://github.com/QuantumNous/new-api/pull/4518
New Contributors
- @Micah-Zheng made their first contribution in https://github.com/QuantumNous/new-api/pull/4518
Full Changelog: https://github.com/QuantumNous/new-api/compare/v1.0.0-alpha.1...v1.0.0-rc.1

New Features
- Model sync now detects upstream models more accurately, refreshes pricing automatically from provider pricing endpoints, shows removed upstream models in the fetch-models dialog, and keeps pricing available during sync delays (#4452).
- User records now include account creation and last login timestamps for easier auditing and administration (#4450).
- Added configurable model matching for Ali native Messages APIs, making it easier to route requests to compatible upstream models (#4468).
Bug Fixes
- Tool calls now accept raw JSON argument payloads, improving compatibility with providers that return tool arguments as JSON objects (#4469).
- Configuration updates now handle JSON object fields more reliably, preventing some settings from being saved incorrectly.
Improvements
- Retry messages now clearly explain when a channel was skipped because affinity is disabled (#4453).

What's Changed
- fix(channel): load model mapping during upstream model checks by @seefs001 in https://github.com/QuantumNous/new-api/pull/4437
- feat: sync upstream pricing from pricing endpoint by @seefs001 in https://github.com/QuantumNous/new-api/pull/4452
- fix: clarify affinity disabled channel retry message by @seefs001 in https://github.com/QuantumNous/new-api/pull/4453
- feat: configure native messages model matching for ali by @seefs001 in https://github.com/QuantumNous/new-api/pull/4468
- fix: support raw JSON response tool arguments by @seefs001 in https://github.com/QuantumNous/new-api/pull/4469
- feat: show removed upstream models in fetch models modal by @seefs001 in https://github.com/QuantumNous/new-api/pull/4470
- 增加用户创建时间和最后登录时间 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4450
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.13.1-patch.1...v0.13.2

New Features
- Added support for handling DeepSeek V4 reasoning suffixes (#4428)
Bug Fixes
- Fixed an issue where tiered billing models were excluded from the model listing (#4431)

What's Changed
- fix: include tiered billing models in model listing by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4431
- feat: implement DeepSeek V4 reasoning suffix handling and tests by @HynoR in https://github.com/QuantumNous/new-api/pull/4428
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.13.0...v0.13.1

Highlights
This patch adds support for using len in tier conditions and the LLM prompt helper, making it easier to build rules based on prompt length.
New Features
- Added a len variable to tier conditions and the LLM prompt helper so you can define rules based on prompt length.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.13.1...v0.13.1-patch.1

Highlights
This release introduces powerful support for expr expression-based tiered billing, allowing administrators to configure complex, dynamic pricing rules. A simple use case for this feature is a pricing expression like tokens <= 128000 ? 0.002 : 0.004, where the system will automatically apply a lower rate for usage under 128K tokens and a higher rate for anything above that threshold.
New Features
- Introduces support for evaluating expr expressions in the billing system, enabling dynamic tiered pricing with configurable presets (#4409).
- Adds a UI for tool pricing settings and improves tool call quota calculations.
- Displays the user ID in admin top-up bills for easier tracking (#4349).
- Adds a 'last used time' column to the tokens table (#4372).
- Relaxes token key column length requirements to support the migration of longer legacy token keys (#4401).
Bug Fixes
- Ensures the pricing calculation logic only applies the N ratio to image models (#4412).
- Preserves text tool surcharges properly during tiered billing settlement (#4162).
- Corrects the completion ratio configuration for the gpt-5.5 model (#4414).
- Fixes the model pricing UI to use the correct display type (#4426).
- Resolves runtime crashes and edge cases within the tiered pricing editor and settlement logic.
- Updates channel testing to properly support and evaluate tiered billing models.
Improvements
- Adds preset pricing tiers for Doubao Seed 1.8 to enhance discount calculations.
- Replaces Card components with divs to improve overall layout consistency.
- Updates internal billing calculations to use quota per unit for better accuracy.

What's Changed
- fix(channel-test): support tiered billing model tests by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4145
- 在管理员充值账单中显示用户 ID by @uskyu in https://github.com/QuantumNous/new-api/pull/4349
- 增加令牌最后使用时间 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4372
- chore(deps): bump github.com/jackc/pgx/v5 from 5.9.0 to 5.9.2 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4397
- chore(deps-dev): bump @xmldom/xmldom from 0.8.12 to 0.8.13 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4399
- Relax token key column length for legacy migration compatibility by @XiaoAI1024 in https://github.com/QuantumNous/new-api/pull/4401
- fix(billing): preserve text tool surcharges in tiered settlement by @yyhhyyyyyy in https://github.com/QuantumNous/new-api/pull/4162
- feat: support for tiered billing expressions in the billing system by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/4409
- fix(image): only price image model use N ratio by @xyfacai in https://github.com/QuantumNous/new-api/pull/4412
- fix: correct gpt-5.5 completion ratio by @jingx8885 in https://github.com/QuantumNous/new-api/pull/4414
- fix: model pricing use correct display type by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4426
New Contributors
- @uskyu made their first contribution in https://github.com/QuantumNous/new-api/pull/4349
- @XiaoAI1024 made their first contribution in https://github.com/QuantumNous/new-api/pull/4401
- @jingx8885 made their first contribution in https://github.com/QuantumNous/new-api/pull/4414
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.15...v0.13.0

New Features
- Added Waffo Pay support for account top-ups and payment workflows (#4089).
Bug Fixes
- Fixed a recharge card issue in the top-up flow that could break parts of the card interface (#4388).
Improvements
- Added support for Gemini's IncludeServerSideToolInvocations setting in ToolConfig, improving compatibility with newer Gemini tool-calling behavior (#4311).
- Improved Codex channel reliability and usability by refreshing credentials for auto-disabled channels, using streaming for automatic tests, and refining the usage modal (#4324).

What's Changed
- fix(gemini): add IncludeServerSideToolInvocations field to ToolConfig by @KoellM in https://github.com/QuantumNous/new-api/pull/4311
- rafactor: payment by @seefs001 in https://github.com/QuantumNous/new-api/pull/4089
- fix(topup): import missing Tag in recharge card by @gaoren002 in https://github.com/QuantumNous/new-api/pull/4388
- chore: refine codex usage modal layout by @seefs001 in https://github.com/QuantumNous/new-api/pull/4386
- fix: use stream for codex auto test by @seefs001 in https://github.com/QuantumNous/new-api/pull/4325
- fix: refresh codex credentials for auto-disabled channels by @seefs001 in https://github.com/QuantumNous/new-api/pull/4324
- fix: require proper verification for passkey changes by @seefs001 in https://github.com/QuantumNous/new-api/pull/4393
New Contributors
- @KoellM made their first contribution in https://github.com/QuantumNous/new-api/pull/4311
- @gaoren002 made their first contribution in https://github.com/QuantumNous/new-api/pull/4388
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.14...v0.12.15

Improvements
- Improved user-visible management and top-up logs with a cleaner display, preserved row expansion, and clearer handling of legacy records.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.13...v0.12.14

This release improves admin workflows for top-ups and makes account status changes apply more consistently.
Bug Fixes
- Disabling a user now takes effect immediately instead of being delayed by stale cached account data.
Improvements
- Top-up logs now include additional admin-only audit details to make review and troubleshooting easier.
Breaking Changes
- Top-up searches are now limited to a maximum 30-day date range to keep queries responsive and reliable.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.12...v0.12.13

This release adds support for Claude Opus 4.7, expanding the set of available models in the API.
New Features
- Added support for the Claude Opus 4.7 model, making it available through the API alongside existing model options (#4293).

What's Changed
- feat: support claude-opus-4-7 by @prnake in https://github.com/QuantumNous/new-api/pull/4293
- chore(deps): bump github.com/jackc/pgx/v5 from 5.7.1 to 5.9.0 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4294
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.11...v0.12.12

New Features
- Added payment method tracking and validation for top-ups.
Bug Fixes
- Improved handling of empty string content during OpenAI to Claude message conversion.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.10...v0.12.11

New Features
- Added passthrough support for Claude cache_control and speed options, giving you more control over request behavior when using Claude models (#4247).
Bug Fixes
- Fixed OpenAI Responses API compatibility issues when instructions contain structured content instead of plain text (#4260).
Improvements
- Improved Stripe payment processing to better handle asynchronous webhook events, making delayed payment confirmations more reliable.
- Quota adjustment logs now record the acting admin username, making administrative audit trails easier to follow (#4216).

What's Changed
- fix: use json.RawMessage for Instructions field in OpenAIResponsesResponse by @linbin524 in https://github.com/QuantumNous/new-api/pull/4260
- fix(channel): 修复多密钥管理弹窗索引显示，将索引值调整为从 1 开始 by @wans10 in https://github.com/QuantumNous/new-api/pull/4231
- feat(claude): add cache_control and speed passthrough controls by @seefs001 in https://github.com/QuantumNous/new-api/pull/4247
- feat: include admin username in quota adjustment logs by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4216
New Contributors
- @linbin524 made their first contribution in https://github.com/QuantumNous/new-api/pull/4260
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.9...v0.12.10

New Features
- Subscription cards now show the next quota reset time, making it easier to see when usage limits will refresh (#4181).
Bug Fixes
- Claude requests are now more reliable by avoiding incompatible Top P settings when needed.
- Azure channels now correctly support the /v1/responses/compact route (#4149).
- Editing values in GroupTable no longer forces the cursor to jump to the end on each keystroke (#4208).
Improvements
- Dashboard charts have been refined to present rankings and chart sizing more clearly for easier monitoring.

What's Changed
- fix(GroupTable): prevent Input cursor jumping to end on keystroke by @Xbang0222 in https://github.com/QuantumNous/new-api/pull/4208
- feat: display next quota reset time in subscription card by @ShengM97 in https://github.com/QuantumNous/new-api/pull/4181
- fix: add Azure channel support for /v1/responses/compact URL routing by @woan1136 in https://github.com/QuantumNous/new-api/pull/4149
New Contributors
- @Xbang0222 made their first contribution in https://github.com/QuantumNous/new-api/pull/4208
- @ShengM97 made their first contribution in https://github.com/QuantumNous/new-api/pull/4181
- @woan1136 made their first contribution in https://github.com/QuantumNous/new-api/pull/4149
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.8...v0.12.9

Bug Fixes
- Correctly report the stream status in error logs instead of defaulting to false (#4195).

What's Changed
- fix: isStream status in error logs instead of hardcoded false by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4195
- chore(deps): bump axios from 1.13.5 to 1.15.0 in /web by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4201
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.7...v0.12.8

Improvements
- Quota management now uses an amount-first input flow and applies changes atomically, making balance adjustments simpler and more reliable.
- Model pricing errors now use cleaner, role-aware messages so it is easier to understand and resolve failed pricing actions.

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.6...v0.12.7

New Features
- Fill in custom fields for vllm-omini (#4154)
Bug Fixes
- Accept string usage values in Alibaba task polling (#4155)
- Use correct endpoint for coding plan image generation in Zhipu 4V (#4146)
- Prefer explicit pricing for compact models (#4156)
- Fix document rendering (#4153)
Improvements
- Refine PR template and add PR submission checks (#4076)

What's Changed
- fix(zhipu_4v): use correct endpoint for coding plan image generation by @NyaMisty in https://github.com/QuantumNous/new-api/pull/4146
- fix: prefer explicit pricing for compact models by @seefs001 in https://github.com/QuantumNous/new-api/pull/4156
- fix(ali): accept string usage values in task polling by @seefs001 in https://github.com/QuantumNous/new-api/pull/4155
- feat: fill in some custom fields for vllm-omini by @seefs001 in https://github.com/QuantumNous/new-api/pull/4154
- fix: document render by @seefs001 in https://github.com/QuantumNous/new-api/pull/4153
- ci: refine PR template and add PR submission checks by @seefs001 in https://github.com/QuantumNous/new-api/pull/4076
New Contributors
- @NyaMisty made their first contribution in https://github.com/QuantumNous/new-api/pull/4146
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.5...v0.12.6

New Features
- Add support for Minimax image generation relay (#4103).
Improvements
- Redesign group ratio rules with a collapsible layout for better usability.
- Enhance dashboard chart axes and sorting logic.
- Add a convenient copy button next to the API link in the dashboard info panel.
Bug Fixes
- Resolve email delivery issues for Outlook and similar providers (#4112).

What's Changed
- feat: 支持强制使用 AUTH LOGIN 以解决 outlook 等邮箱的发件问题 by @Hoshino-Yumetsuki in https://github.com/QuantumNous/new-api/pull/4112
- feat(minimax): add image generation relay support by @forsakenyang in https://github.com/QuantumNous/new-api/pull/4103
New Contributors
- @Hoshino-Yumetsuki made their first contribution in https://github.com/QuantumNous/new-api/pull/4112
- @forsakenyang made their first contribution in https://github.com/QuantumNous/new-api/pull/4103
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.4...v0.12.5

New Features
- Added an IncludeModelName option to channel affinity rules for per-model affinity tracking (#3488)
- Added an ErrorBoundary component to the web UI to prevent full-page crashes (#3350)
- Added support for Duration in TaskSubmitReq (#4068)
- Added admin user analytics to the dashboard
Bug Fixes
- Fixed pricing filtering based on usable groups (#4123)
- Fixed an issue where Claude streaming interruptions incorrectly overwrote usage data instead of preserving cache billing fields (#4128)
- Fixed an issue where the "do not retry after failure" configuration was incorrectly overwritten in memory (#4142)
Improvements
- Added support for enabling error logging via environment variables (#4131)
- Wrapped scope tag labels with t() for better i18n support
- Fixed chart labels on the dashboard

What's Changed
- feat: add IncludeModelName option to channel affinity rules by @clansty in https://github.com/QuantumNous/new-api/pull/3488
- feat: add ErrorBoundary to prevent full-page crashes by @goodmorning10 in https://github.com/QuantumNous/new-api/pull/3350
- fix: Claude 流式断流时不再整份覆盖 usage，保留 cache 计费字段 by @zuiho-kai in https://github.com/QuantumNous/new-api/pull/4128
- fix(pricing): add filtering for pricing based on usable groups by @bbbugg in https://github.com/QuantumNous/new-api/pull/4123
- chore(deps): bump github.com/aws/aws-sdk-go-v2/service/bedrockruntime from 1.50.0 to 1.50.4 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4136
- add 添加 启用错误日志记录到 env 配置中 by @binorxin in https://github.com/QuantumNous/new-api/pull/4131
- Seedance support duration by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4068
- fix: 修复 失败后不重试 配置项写到内存被覆盖 by @seefs001 in https://github.com/QuantumNous/new-api/pull/4142
New Contributors
- @clansty made their first contribution in https://github.com/QuantumNous/new-api/pull/3488
- @goodmorning10 made their first contribution in https://github.com/QuantumNous/new-api/pull/3350
- @zuiho-kai made their first contribution in https://github.com/QuantumNous/new-api/pull/4128
- @bbbugg made their first contribution in https://github.com/QuantumNous/new-api/pull/4123
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.3...v0.12.4

This is a small polish release focused on making configuration easier to navigate by simplifying the settings layout and consolidating model pricing controls.
Improvements
- Refined the settings interface with a unified model pricing view and a cleaner tab structure to make configuration easier to navigate.

What's Changed
- feat(token): add batch API for fetching token keys by @RedwindA in https://github.com/QuantumNous/new-api/pull/4114
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.2...v0.12.3

New Features
- Add support for PDF conversion between OpenAI and Claude APIs.
- Support differential billing for Seedance 2.0 based on video vs. no-video input.
- Enhance max_tokens handling and input sanitization in the playground (#4106).
- Enhance stream status display with error tooltips in usage logs.
Bug Fixes
- Fix usage calculation issues when converting between Claude and OpenAI APIs by emitting message_delta for the final stream chunk (#4090).
- Fix Gemini stream detection by checking the :streamGenerateContent URL path (#4087).
- Fix Claude request requirements by properly setting TopP to nil.
Improvements
- Unify file source creation and enhance caching mechanisms.

What's Changed
- fix(gemini): detect streaming from URL path 😒treamGenerateContent by @D26FORWARD in https://github.com/QuantumNous/new-api/pull/4087
- fix: emit claude message_delta for usage-only final stream chunk by @seefs001 in https://github.com/QuantumNous/new-api/pull/4090
- chore(deps-dev): bump electron from 35.7.5 to 39.8.5 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4080
- chore(deps-dev): bump @xmldom/xmldom from 0.8.11 to 0.8.12 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/3523
- chore(deps-dev): bump lodash from 4.17.23 to 4.18.1 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/4097
- feat(playground): enhance max_tokens handling and input sanitization by @HynoR in https://github.com/QuantumNous/new-api/pull/4106
New Contributors
- @D26FORWARD made their first contribution in https://github.com/QuantumNous/new-api/pull/4087
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.1...v0.12.2

This release adds Seedance 2.0 video support and broadens media compatibility across providers. It also makes failures easier to understand with clearer Seedance and performance-related error messages.
New Features
- Added Seedance 2.0 video generation API support with duration control in seconds, clearer failure reporting, and better prompt handling for single-text requests (#4042).
- Added HEIC/HEIF image support, including Gemini channel compatibility and more reliable handling of HEIF images (#4049).
- Added basic inline file support for Claude relay so compatible requests can include media directly (#3505).
Improvements
- Improved performance-related error messages to make request failures easier to understand.

What's Changed
- fix: add basic inline file support for Claude relay by @seefs001 in https://github.com/QuantumNous/new-api/pull/3505
- 新增 seedance2.0 视频接口支持 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/4042
- feat: add HEIC/HEIF image format support for Gemini channel by @RedwindA in https://github.com/QuantumNous/new-api/pull/4049
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.12.0...v0.12.1

Release Highlights
This release introduces support for the wan2.7 image generation model and adds persistence for advanced settings in the EditChannelModal. It also includes several UI bug fixes and billing logic corrections.
New Features
- Models: Add support for wan 2.7 image generation (#3526)
- UI: Persist advanced settings state in local storage within EditChannelModal
Bug Fixes
- Billing: Prevent double-counting of image count n by moving it to OtherRatio (#3512)
- Dashboard: Fix scrollbar flickering when hovering over the consumption distribution chart (#3474)
- xAI: Set MaxTokens to nil when MaxCompletionTokens is 0 for the grok-3-mini model
Improvements
- UI: Enhance footer layout and add custom classes for styling
- UI: Enhance clipboard handling with error checks in EditChannelModal
- Docs: Update Traditional Chinese README with missing content and partner logo (#3462)

What's Changed
- fix(dashboard): 修复消耗分布图表悬浮时滚动条闪烁 by @wans10 in https://github.com/QuantumNous/new-api/pull/3474
- docs(zh-TW): fix missing content and add partner logo by @DaZuiZui in https://github.com/QuantumNous/new-api/pull/3462
- chore(deps-dev): bump picomatch from 4.0.3 to 4.0.4 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/3450
- fix: prevent double-counting of image count n in billing by @imlhb in https://github.com/QuantumNous/new-api/pull/3512
- 支持 wan2.7 生图-wan2.7-image by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3526
New Contributors
- @DaZuiZui made their first contribution in https://github.com/QuantumNous/new-api/pull/3462
- @imlhb made their first contribution in https://github.com/QuantumNous/new-api/pull/3512
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.9...v0.12.0

What's Changed
- fix: disable doubao coding plan selection by @seefs001 in https://github.com/QuantumNous/new-api/pull/3379
- fix: oauth bind callback handling by @seefs001 in https://github.com/QuantumNous/new-api/pull/3393
- feat: support regex-prefixed ignored upstream models by @seefs001 in https://github.com/QuantumNous/new-api/pull/3381
- adjuct default settings by @seefs001 in https://github.com/QuantumNous/new-api/pull/3335
- fix: honor channel affinity skip-retry when channel is disabled by @seefs001 in https://github.com/QuantumNous/new-api/pull/3333
- fix: apply forced beta query at final upstream URL stage by @seefs001 in https://github.com/QuantumNous/new-api/pull/3331
- Refactor/codex usage by @seefs001 in https://github.com/QuantumNous/new-api/pull/3399
- refactor: optimize billing flow for OpenAI-to-Anthropic convert by @seefs001 in https://github.com/QuantumNous/new-api/pull/3400
- fix: the "detail" field is empty, an empty field was sent to upstream by @seefs001 in https://github.com/QuantumNous/new-api/pull/3401
- fix: restore pre-3400 OpenRouter billing semantics by @seefs001 in https://github.com/QuantumNous/new-api/pull/3438
- refactor: expose skip-retry option and show it in rules list by @seefs001 in https://github.com/QuantumNous/new-api/pull/3440
- feat: prevent metadata from overriding model fields by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3461
- chore(deps): bump golang.org/x/image from 0.23.0 to 0.38.0 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/3507
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.8...v0.11.9

What's Changed
- fix: normalize generic oauth bearer token type by @seefs001 in https://github.com/QuantumNous/new-api/pull/3359
- feat: add server log file management to performance settings by @RedwindA in https://github.com/QuantumNous/new-api/pull/3369
- feat: Add support for counting cache-hit tokens in llama.cpp by @wenyifancc in https://github.com/QuantumNous/new-api/pull/3357
- docs: 完善宝塔面板部署教程并修复链接错误 by @lcq225 in https://github.com/QuantumNous/new-api/pull/3360
New Contributors
- @wenyifancc made their first contribution in https://github.com/QuantumNous/new-api/pull/3357
- @lcq225 made their first contribution in https://github.com/QuantumNous/new-api/pull/3360
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.7...v0.11.8

What's Changed
- chore: refine PR template by @seefs001 in https://github.com/QuantumNous/new-api/pull/3287
- feat: 支持通过环境变量配置搜索接口限流参数 by @ywandy in https://github.com/QuantumNous/new-api/pull/3313
- fix: 修正 Codex free 账号用量显示到每周窗口 by @Honghurumeng in https://github.com/QuantumNous/new-api/pull/3316
- feat(waffo): Waffo payment gateway integration by @zhongyuanzhao-alt in https://github.com/QuantumNous/new-api/pull/3293
- fix: redirect OAuth login in current page by @seefs001 in https://github.com/QuantumNous/new-api/pull/3329
New Contributors
- @ywandy made their first contribution in https://github.com/QuantumNous/new-api/pull/3313
- @Honghurumeng made their first contribution in https://github.com/QuantumNous/new-api/pull/3316
- @zhongyuanzhao-alt made their first contribution in https://github.com/QuantumNous/new-api/pull/3293
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.6...v0.11.7

参数覆盖记录到日志
错误日志恢复 tooltip 显示
What's Changed
- feat: params override log by @seefs001 in https://github.com/QuantumNous/new-api/pull/3285
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.5...v0.11.6

What's Changed
- Round remaining balance by @KiGamji in https://github.com/QuantumNous/new-api/pull/3233
- enhance channel key viewing by @seefs001 in https://github.com/QuantumNous/new-api/pull/3257
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.4...v0.11.5

What's Changed
- Feature/param override wildcard path by @seefs001 in https://github.com/QuantumNous/new-api/pull/3129
- fix: fetch model add header passthrough rule key check by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3130
- feats: repair the thinking of claude to openrouter convert by @nekohy in https://github.com/QuantumNous/new-api/pull/3120
- feat: kling cost quota support use FinalUnitDeduction as totalToken by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2769
- fix: If top_p is not provided, ignore it by @seefs001 in https://github.com/QuantumNous/new-api/pull/3141
- fix(relay): skip retries for bad response body errors by @seefs001 in https://github.com/QuantumNous/new-api/pull/3151
- 为渠道参数覆盖可视化规则提供拖拽排序支持 by @somnifex in https://github.com/QuantumNous/new-api/pull/3166
- fix: kling risk fail return openAIVideo error by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3148
- fix: add explicit docker-compose networks by @pigletfly in https://github.com/QuantumNous/new-api/pull/3147
- feat：support $keep_only_declared and deduped $append for header override by @seefs001 in https://github.com/QuantumNous/new-api/pull/3182
- chore: update model lists for frequently used channels by @RedwindA in https://github.com/QuantumNous/new-api/pull/3221
New Contributors
- @pigletfly made their first contribution in https://github.com/QuantumNous/new-api/pull/3147
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.2...v0.11.4

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.4...v0.11.4-patch.1

Main changes
增加定时获取上游模型更新功能
异步任务在没有设置价格时抛出错误
渠道参数覆盖功能增强，提供可视化编辑，整合请求头覆盖和透传功能
Add a scheduled feature to fetch upstream model updates.
Throw an error in the asynchronous task when the price is not set.
Enhance the channel parameter override feature by integrating request header overrides and passthrough functionality.
Enhance the channel parameter override feature
整合请求头覆盖功能，可视化编辑，并提供基础模板
Scheduled feature to fetch upstream model updates
在上游模型变动时，可看到明显的提醒
点击后可分别处理需要删除和新增的模型
What's Changed
- fix: redirect subscription payment return to user-accessible page by @seefs001 in https://github.com/QuantumNous/new-api/pull/3052
- feat: add image model to supported image presets by @seefs001 in https://github.com/QuantumNous/new-api/pull/3029
- feat: minimax reasoning_split by @seefs001 in https://github.com/QuantumNous/new-api/pull/3012
- feat(gemini): implement video generation configuration by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/3059
- feat: improve channel override ui/ux by @seefs001 in https://github.com/QuantumNous/new-api/pull/3009
- 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/3061
- chore(deps): bump tar and electron-builder in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2964
- build(deps-dev): bump lodash from 4.17.21 to 4.17.23 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2720
- chore(deps-dev): bump minimatch from 3.1.2 to 3.1.5 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/3060
- fix: preserve explicit zero values in native relay requests by @seefs001 in https://github.com/QuantumNous/new-api/pull/3069
- fix: aws text content blocks must be non-empty by @seefs001 in https://github.com/QuantumNous/new-api/pull/3077
- Fix/aws header override by @seefs001 in https://github.com/QuantumNous/new-api/pull/3066
- Fix/aws non empty text by @seefs001 in https://github.com/QuantumNous/new-api/pull/3080
- Revert "Fix/aws non empty text" by @seefs001 in https://github.com/QuantumNous/new-api/pull/3082
- Revert "fix: aws text content blocks must be non-empty" by @seefs001 in https://github.com/QuantumNous/new-api/pull/3083
- fix: change token model_limits column from varchar(1024) to text by @RedwindA in https://github.com/QuantumNous/new-api/pull/3037
- Return error when model price/ratio unset by @BenLampson in https://github.com/QuantumNous/new-api/pull/3081
- feat: auto fetch upstream models by @seefs001 in https://github.com/QuantumNous/new-api/pull/2979
- fix: use default model price for radio price model by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3090
- Fix/auto fetch upstream model tips by @seefs001 in https://github.com/QuantumNous/new-api/pull/3096
- fix: update task billing log content to include reason by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3093
New Contributors
- @BenLampson made their first contribution in https://github.com/QuantumNous/new-api/pull/3081
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.0...v0.11.2

优化 kling 模型计费
What's Changed
- feat: kling cost quota support use FinalUnitDeduction as totalToken by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2769
- fix: If top_p is not provided, ignore it by @seefs001 in https://github.com/QuantumNous/new-api/pull/3141
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.2-patch.1...v0.11.2-patch.2

gpt-5.4 补全倍率
What's Changed
- Feature/param override wildcard path by @seefs001 in https://github.com/QuantumNous/new-api/pull/3129
- fix: fetch model add header passthrough rule key check by @feitianbubu in https://github.com/QuantumNous/new-api/pull/3130
- feats: repair the thinking of claude to openrouter convert by @nekohy in https://github.com/QuantumNous/new-api/pull/3120
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.11.2...v0.11.2-patch.1

Major Changes
- Channel parameter override now supports copy, trim_prefix, trim_suffix, ensure_prefix, ensure_suffix, trim_space, to_lower, to_upper, replace, and regex_replace operations.
- Doubao now supports the native /v1/responses format.
- io.net i18n support, and deployment operation UI/UX optimization.
- Check-in now supports integration with Cloudflare Turnstile.
主要变更
- 渠道参数覆盖增加 copy, trim_prefix, trim_suffix, ensure_prefix, ensure_suffix, trim_space, to_lower, to_upper, replace, and regex_replace 操作
- 豆包支持 /v1/responses 原生格式
- io.net i18n 支持，部署操作 UI/UX 优化
- 签到支持接入 Cloudflare Turnstile
What's Changed
- fix(checkin): prevent visual flicker when loading check-in component by @RedwindA in https://github.com/QuantumNous/new-api/pull/2566
- remove duplicate condition in TaskLogsColumnDefs by @zpc7 in https://github.com/QuantumNous/new-api/pull/2554
- feat: 支持仅使用 x-api-key 获取 anthropic 格式的模型列表 #2422 by @atopos31 in https://github.com/QuantumNous/new-api/pull/2425
- feat: support HTTP_PROXY environment variable for default HTTP client by @wwalt1a in https://github.com/QuantumNous/new-api/pull/2547
- fix: kling correct fail reason by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2532
- feat: support claude-haiku-4-5-20251001 on vertex by @prnake in https://github.com/QuantumNous/new-api/pull/2393
- fix(task): 修复渠道配置多个 key 时无法获取任务的问题 by @a4399518s in https://github.com/QuantumNous/new-api/pull/2447
- fix: CrossGroupRetry default false by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2570
- feat: add support for Doubao /v1/responses by @seefs001 in https://github.com/QuantumNous/new-api/pull/2567
- feat: check-in feature integrates Turnstile security check by @seefs001 in https://github.com/QuantumNous/new-api/pull/2571
- fix: gemini request → openai tool call by @seefs001 in https://github.com/QuantumNous/new-api/pull/2558
- fix: fix model deployment style issues, lint problems, and i18n gaps. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2556
- feat: add parameter coverage for more operations by @seefs001 in https://github.com/QuantumNous/new-api/pull/2568
New Contributors
- @zpc7 made their first contribution in https://github.com/QuantumNous/new-api/pull/2554
- @atopos31 made their first contribution in https://github.com/QuantumNous/new-api/pull/2425
- @wwalt1a made their first contribution in https://github.com/QuantumNous/new-api/pull/2547
- @a4399518s made their first contribution in https://github.com/QuantumNous/new-api/pull/2447
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.5-checkin.1...v0.10.5-ionet.2

What's Changed
- feat: map OpenAI developer role to Gemini system instructions by @RedwindA in https://github.com/QuantumNous/new-api/pull/2536
- feat: ionet integrate by @seefs001 in https://github.com/QuantumNous/new-api/pull/2105
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.4...v0.10.5-ionet.1

What's Changed
- feat: pyroscope integrate by @seefs001 in https://github.com/QuantumNous/new-api/pull/2475
- fix: ensure overwrite works correctly when no missing models by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2154
- fix(i18n): disable namespace separator to fix URL display in translations by @RedwindA in https://github.com/QuantumNous/new-api/pull/2530
- 🚀 fix(model-sync): avoid unnecessary upstream fetch while keeping ove… by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2528
- fix: 支持小写 bearer 和 Bearer 后带多个空格 && 修复 WSS 预扣费错误提取 key 的问题 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2513
- fix: add warning for pass through body by @seefs001 in https://github.com/QuantumNous/new-api/pull/2512
- fix: 支持传入 system_instruction 和 systemInstruction 两种风格系统提示词参数名 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2450
- Refine French translations for UI conciseness by @comeback01 in https://github.com/QuantumNous/new-api/pull/2455
- fix(gemini): handle minimal reasoning effort budget by @seefs001 in https://github.com/QuantumNous/new-api/pull/2460
- fix: Fix Openrouter test errors and optimize error messages by @seefs001 in https://github.com/QuantumNous/new-api/pull/2433
- feat: support first bind update password by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2520
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.3...v0.10.4

What's Changed
- 🔗 docs(readme): update documentation links to new site routing by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2486
- 修复 claudeResponse 流式请求空指针 Panic by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2510
- Solve Claude Code crash issue by @JerryKwan in https://github.com/QuantumNous/new-api/pull/2511
- 为 Moonshot 添加缓存 tokens 读取逻辑 by @shikaiwei1 in https://github.com/QuantumNous/new-api/pull/2493
- fix: 修复 Anthropic 渠道缓存计费错误 by @1420970597 in https://github.com/QuantumNous/new-api/pull/2477
- fix: revert model ratio by @seefs001 in https://github.com/QuantumNous/new-api/pull/2524
New Contributors
- @JerryKwan made their first contribution in https://github.com/QuantumNous/new-api/pull/2511
- @shikaiwei1 made their first contribution in https://github.com/QuantumNous/new-api/pull/2493
- @1420970597 made their first contribution in https://github.com/QuantumNous/new-api/pull/2477
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.2...v0.10.3

What's Changed
- fix: 模型设置增加针对 Vertex 渠道过滤 content[].part[].functionResponse.id 的选项，默认启用 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2483
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.1...v0.10.2

主要更新
- 支持在令牌中设置 auto 分组下是否启用跨分组重试
- 支持 gpt 模型-xhigh 后缀
- 支持非流 gpt-4o-mini-tts 类模型计费，规则详见 https://github.com/QuantumNous/new-api/pull/2434
What's Changed
- fix: 支持 aws 通过全局参数透传或者渠道参数透传来 调用 by @zdwy5 in https://github.com/QuantumNous/new-api/pull/2423
- fix: health check by @seefs001 in https://github.com/QuantumNous/new-api/pull/2428
- feat(token): add cross-group retry option for token processing by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2426
- fix: correct sender format issues fix #1347 by @ion1ze in https://github.com/QuantumNous/new-api/pull/2424
- feat(adaptor): add '-xhigh' suffix to reasoning effort options by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2429
- fix(channel_select): adjust priority retry logic for cross-group by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2430
- feat: support gpt tts series model quota calculate by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2434
- feat(auth): enhance IP restriction handling with CIDR support by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2445
- 🛡️ fix: prevent OOM on large/decompressed requests; skip heavy prompt… by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2452
- style: add card spacing by @TinsFox in https://github.com/QuantumNous/new-api/pull/2474
- chore: add code-inspector-plugin integration by @TinsFox in https://github.com/QuantumNous/new-api/pull/2476
New Contributors
- @zdwy5 made their first contribution in https://github.com/QuantumNous/new-api/pull/2423
- @ion1ze made their first contribution in https://github.com/QuantumNous/new-api/pull/2424
- @TinsFox made their first contribution in https://github.com/QuantumNous/new-api/pull/2474
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.10.0...v0.10.1

主要更新
- 支持 sora remix
- 修复可能导致 CPU 占用 100%的问题
- 性能优化
- 优化 OpenAI 转 Claude 格式函数调用的问题
What's Changed
- chore: update the relay openapi file by @t0ng7u in https://github.com/QuantumNous/new-api/pull/2353
- fix: try resolve the high concurrency issue to a single host by @prnake in https://github.com/QuantumNous/new-api/pull/2351
- feat: refactor token estimation logic by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2355
- feat: gemini 3 thinking level gemini-3-pro-preview-high by @seefs001 in https://github.com/QuantumNous/new-api/pull/2344
- fix(adaptor): fix reasoning suffix not processing in vertex adapter by @FirstMelody in https://github.com/QuantumNous/new-api/pull/2388
- feat: add claude-haiku-4-5-20251001 model support by @FlowerRealm in https://github.com/QuantumNous/new-api/pull/2375
- Increase token name length limit from 30 to 50 by @oudi in https://github.com/QuantumNous/new-api/pull/2368
- fix: Set default to unsupported value for gpt-5 model series requests by @HynoR in https://github.com/QuantumNous/new-api/pull/2229
- fix(go.mod): 更新 modernc.org/sqlite 依赖项版本 by @binorxin in https://github.com/QuantumNous/new-api/pull/2387
- feat: zhipu 4v image generations by @seefs001 in https://github.com/QuantumNous/new-api/pull/2356
- chore(go): enable greenteagc by @seefs001 in https://github.com/QuantumNous/new-api/pull/2357
- fix: regex repeat compile by @seefs001 in https://github.com/QuantumNous/new-api/pull/2358
- fix: qwen chat_template_kwargs by @seefs001 in https://github.com/QuantumNous/new-api/pull/2359
- fix: fetch upstream models by @seefs001 in https://github.com/QuantumNous/new-api/pull/2394
- fix: sidebar color overlap by @seefs001 in https://github.com/QuantumNous/new-api/pull/2395
- fix: Try to fix login error "already logged in" issue by @seefs001 in https://github.com/QuantumNous/new-api/pull/2396
- fix: Use channel proxy settings for task query scenarios by @seefs001 in https://github.com/QuantumNous/new-api/pull/2398
- feat: update price display use current currency symbol by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2360
- fix: try to fix tool call issues by @seefs001 in https://github.com/QuantumNous/new-api/pull/2397
- fix(channel): update channel identification logic in error processing by @NoahCodeGG in https://github.com/QuantumNous/new-api/pull/2194
- feat: add openai video remix endpoint by @seefs001 in https://github.com/QuantumNous/new-api/pull/2412
New Contributors
- @FirstMelody made their first contribution in https://github.com/QuantumNous/new-api/pull/2388
- @FlowerRealm made their first contribution in https://github.com/QuantumNous/new-api/pull/2375
- @oudi made their first contribution in https://github.com/QuantumNous/new-api/pull/2368
- @NoahCodeGG made their first contribution in https://github.com/QuantumNous/new-api/pull/2194
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.28...v0.10.0

What's Changed
- feat(gemini): implement markdown image handling in text processing by @Calcium-Ion in https://github.com/QuantumNous/new-api/pull/2346
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.27...v0.9.28

What's Changed
- fix: cast size to int64 before comparing with MaxUint32 by @binorxin in https://github.com/QuantumNous/new-api/pull/2264
- fix: root page does not have analytic code by @wzxjohn in https://github.com/QuantumNous/new-api/pull/2261
- feat: add claude-opus-4-5-20251101 by @prnake in https://github.com/QuantumNous/new-api/pull/2293
- feat: enable thoughtSignature for non-function-call messages by @amikebzek in https://github.com/QuantumNous/new-api/pull/2282
- fix: volcengine && baidu claude adapter by @seefs001 in https://github.com/QuantumNous/new-api/pull/2294
- feat: 视频下载和界面预览统一使用 OAI 标准接口 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2324
- fix(aws): extract HTTP status code from AWS SDK errors by @mfzzf in https://github.com/QuantumNous/new-api/pull/2329
- feat(playground): enhance SSE debugging and add image paste support with i18n by @ImogeneOctaviap794 in https://github.com/QuantumNous/new-api/pull/2312
- fix: nano banana pro 4k(StreamScannerMaxBufferMB env) by @seefs001 in https://github.com/QuantumNous/new-api/pull/2335
- feat: glm coding plan && kimi coding plan by @seefs001 in https://github.com/QuantumNous/new-api/pull/2334
- fix: edit vertex key type by @seefs001 in https://github.com/QuantumNous/new-api/pull/2311
- fix: claude request missing field by @seefs001 in https://github.com/QuantumNous/new-api/pull/2304
- fix(i18n): fill missing translations in i18n. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2314
New Contributors
- @binorxin made their first contribution in https://github.com/QuantumNous/new-api/pull/2264
- @amikebzek made their first contribution in https://github.com/QuantumNous/new-api/pull/2282
- @mfzzf made their first contribution in https://github.com/QuantumNous/new-api/pull/2329
- @ImogeneOctaviap794 made their first contribution in https://github.com/QuantumNous/new-api/pull/2312
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.25...v0.9.27-patch.1

What's Changed
- fix: 错误解析 responses api 中的 input 字段 by @jarvis-u in https://github.com/QuantumNous/new-api/pull/2224
- feat: Add Vietnamese language support by @chokiproai in https://github.com/QuantumNous/new-api/pull/2268
- feat: 关联 discord 账号 by @StageDog in https://github.com/QuantumNous/new-api/pull/1706
- feat: gemini thinking_level && snake params by @seefs001 in https://github.com/QuantumNous/new-api/pull/2274
- feat: embedding param override && internal params by @seefs001 in https://github.com/QuantumNous/new-api/pull/2276
- feat: 二次确认添加重定向前模型 && 重定向后模式视为已有模型 by @seefs001 in https://github.com/QuantumNous/new-api/pull/2277
- fix: release workflow show version by @seefs001 in https://github.com/QuantumNous/new-api/pull/2278
New Contributors
- @jarvis-u made their first contribution in https://github.com/QuantumNous/new-api/pull/2224
- @chokiproai made their first contribution in https://github.com/QuantumNous/new-api/pull/2268
- @StageDog made their first contribution in https://github.com/QuantumNous/new-api/pull/1706
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.24...v0.9.25

What's Changed
- feat: Fill thoughtSignature only for Gemini/Vertex channels using OpenAI format by @seefs001 in https://github.com/QuantumNous/new-api/pull/2256
- fix: When retrieving the model list with multiple keys, select the first enabled one. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2260
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.23...v0.9.24

What's Changed
- update model categories' match rules by @QAbot-zh in https://github.com/QuantumNous/new-api/pull/2239
- fix: optimized the GitHub login copy and timeout. by @seefs001 in https://github.com/QuantumNous/new-api/pull/2244
- chore(deps-dev): bump js-yaml from 4.1.0 to 4.1.1 in /electron by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2231
- feat: gemini-3-pro by @seefs001 in https://github.com/QuantumNous/new-api/pull/2243
- feat: channel by tag omit key by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2247
- chore(deps): bump golang.org/x/crypto from 0.42.0 to 0.45.0 by @dependabot[bot] in https://github.com/QuantumNous/new-api/pull/2252
- fix: claude cache price render by @seefs001 in https://github.com/QuantumNous/new-api/pull/2250
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.22...v0.9.23

What's Changed
- 增加 wan2.5-i2i-preview 图生图支持 by @feitianbubu in https://github.com/QuantumNous/new-api/pull/2227
- support reasoning field for playground by @QAbot-zh in https://github.com/QuantumNous/new-api/pull/2207
- feat: support gpt-5.1 prompt_cache_retention by @seefs001 in https://github.com/QuantumNous/new-api/pull/2228
- fix: boundary parser error (error parsing multipart NextPart: bufio: buffer full) by @seefs001 in https://github.com/QuantumNous/new-api/pull/2235
- feat: support vertex open source models by @seefs001 in https://github.com/QuantumNous/new-api/pull/2236
- feat: support configuring the linuxdo endpoint via environment variables by @seefs001 in https://github.com/QuantumNous/new-api/pull/2237
- feat: support doubao coding plan by @seefs001 in https://github.com/QuantumNous/new-api/pull/2238
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.21...v0.9.22

What's Changed
- update model categories' match rules by @QAbot-zh in https://github.com/QuantumNous/new-api/pull/2239
Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.22...v0.9.22-patch.1

Full Changelog: https://github.com/QuantumNous/new-api/compare/v0.9.20-patch.1...v0.9.20-patch.2