💄 style: modelParse support to get model settings params & Customize extendParams UI #11185

sxjeru · 2026-01-04T05:10:26Z

💻 Change Type

🔗 Related Issue

🔀 Description of Change

继承自 #9442

模型抓取允许获取本地配置已有的 extendParams 和 searchSetting
searchSetting 限定仅 aihubmix、newapi 等透传 providers 可抓取获得
extendParams 限定仅透传 providers 可抓取获得所有参数，否则仅 openai/gemini 系模型参数
模型配置界面支持修改 extendParams
重构 extendParams 元件

主要效果，让抓取的模型也支持附加参数功能（和模型内置搜索功能），也允许自行调整单个模型的附加参数功能。

UI 预览：

🧪 How to Test

Tested locally
Added/updated tests
No tests needed

📸 Screenshots / Videos

Before	After
...	...

📝 Additional Information

…code reusability and maintainability. Updated GPT52ProReasoningEffortSlider, GPT52ReasoningEffortSlider, GPT5ReasoningEffortSlider, ImageAspectRatioSelect, ImageResolutionSlider, ReasoningEffortSlider, TextVerbositySlider, ThinkingLevel2Slider, ThinkingLevelSlider, and ThinkingSlider to implement controlled and uncontrolled modes. Enhanced aspect ratio selection and resolution sliders with better type safety and default values.

…tory function for better code reuse and maintainability - Created `createLevelSliderComponent` to handle both controlled and uncontrolled modes for sliders. - Updated `GPT5ReasoningEffortSlider`, `GPT51ReasoningEffortSlider`, `GPT52ProReasoningEffortSlider`, `GPT52ReasoningEffortSlider`, `ImageResolutionSlider`, `ImageAspectRatioSelect`, `ReasoningEffortSlider`, `TextVerbositySlider`, `ThinkingSlider`, `ThinkingLevelSlider`, and `ThinkingLevel2Slider` to utilize the new factory function. - Simplified the logic for handling controlled and uncontrolled states in sliders. - Added tests for the new slider factory to ensure correct behavior in both controlled and uncontrolled modes.

vercel · 2026-01-04T05:10:29Z

@sxjeru is attempting to deploy a commit to the LobeHub OSS Team on Vercel.

A member of the Team first needs to authorize it.

sourcery-ai

Sorry @sxjeru, you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

codecov · 2026-01-04T05:25:52Z

Codecov Report

❌ Patch coverage is 74.82014% with 70 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.90%. Comparing base (b3690cb) to head (5877cd5).
⚠️ Report is 5 commits behind head on next.

Additional details and impacted files

@@            Coverage Diff             @@
##             next   #11185      +/-   ##
==========================================
- Coverage   75.91%   75.90%   -0.02%     
==========================================
  Files        1143     1145       +2     
  Lines       88942    89187     +245     
  Branches     9957    12009    +2052     
==========================================
+ Hits        67519    67694     +175     
- Misses      21345    21404      +59     
- Partials       78       89      +11

Flag	Coverage Δ
app	`69.57% <86.27%> (+0.03%)`	⬆️
database	`93.59% <100.00%> (+<0.01%)`	⬆️
packages/agent-runtime	`89.08% <ø> (ø)`
packages/context-engine	`83.19% <ø> (ø)`
packages/conversation-flow	`92.41% <ø> (ø)`
packages/file-loaders	`88.66% <ø> (ø)`
packages/memory-user-memory	`70.07% <ø> (ø)`
packages/model-bank	`100.00% <ø> (ø)`
packages/model-runtime	`86.59% <67.81%> (-0.30%)`	⬇️
packages/prompts	`77.43% <ø> (ø)`
packages/python-interpreter	`92.90% <ø> (ø)`
packages/ssrf-safe-fetch	`0.00% <ø> (ø)`
packages/utils	`93.20% <ø> (ø)`
packages/web-crawler	`95.62% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
Store	`67.53% <ø> (ø)`
Services	`51.57% <ø> (ø)`
Server	`73.63% <ø> (ø)`
Libs	`41.53% <ø> (ø)`
Utils	`93.51% <ø> (ø)`

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

gru-agent · 2026-01-04T06:40:52Z

There is too much information in the pull request to test.

sourcery-ai

Sorry @sxjeru, you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

Copilot

Pull request overview

This PR enhances model configuration capabilities by enabling fetching and customization of extendParams and searchSetting for models. It introduces a factory pattern to refactor slider components and adds a comprehensive UI for managing extended parameters in the model configuration interface.

Refactored slider components to use a reusable factory pattern with createLevelSliderComponent
Added support for fetching extendParams and searchSetting from local model configurations
Implemented ExtendParamsSelect component with preview functionality for model settings
Enhanced provider implementations (vercelaigateway, openrouter) to automatically detect and configure extended parameters

Reviewed changes

Copilot reviewed 33 out of 33 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
src/locales/default/modelProvider.ts	Added translations for extended parameters options and their descriptions
src/locales/default/chat.ts	Added translation for "Thinking Budget" title
src/features/ChatInput/ActionBar/Model/createLevelSlider.tsx	New factory function to create level slider components with controlled/uncontrolled modes
src/features/ChatInput/ActionBar/Model/tests/createLevelSlider.test.tsx	Tests verifying factory pattern works in controlled mode without store access
src/features/ChatInput/ActionBar/Model/ThinkingSlider.tsx	Refactored to use factory pattern, reducing boilerplate
src/features/ChatInput/ActionBar/Model/ThinkingLevelSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/ThinkingLevel2Slider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/TextVerbositySlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/ReasoningEffortSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/LevelSlider.tsx	New generic level slider component used by factory
src/features/ChatInput/ActionBar/Model/ImageResolutionSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/ImageAspectRatioSelect.tsx	Refactored to support controlled/uncontrolled modes for preview
src/features/ChatInput/ActionBar/Model/GPT5ReasoningEffortSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/GPT52ReasoningEffortSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/GPT52ProReasoningEffortSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/GPT51ReasoningEffortSlider.tsx	Refactored to use factory pattern
src/features/ChatInput/ActionBar/Model/ControlsForm.tsx	Fixed translation key for thinking budget
src/app/.../CreateNewModelModal/tests/ExtendParamsSelect.test.tsx	Tests for title key aliases mapping logic
src/app/.../CreateNewModelModal/Form.tsx	Added ExtendParamsSelect to model configuration form
src/app/.../CreateNewModelModal/ExtendParamsSelect.tsx	New component for selecting and previewing extended parameters
packages/types/src/llm.ts	Added settings field to ChatModelCard interface
packages/model-runtime/src/utils/modelParse.ts	Added logic to merge extendParams and searchSettings based on provider type
packages/model-runtime/src/utils/modelParse.test.ts	Tests for settings merging behavior
packages/model-runtime/src/providers/vercelaigateway/index.ts	Updated payload handling and added automatic extendParams detection
packages/model-runtime/src/providers/vercelaigateway/index.test.ts	Updated tests for new model prefix format
packages/model-runtime/src/providers/openrouter/type.ts	Refactored OpenRouterReasoning interface
packages/model-runtime/src/providers/openrouter/index.ts	Updated reasoning handling and added automatic extendParams detection
packages/model-runtime/src/providers/openrouter/index.test.ts	Updated tests for new reasoning behavior
packages/database/src/repositories/aiInfra/index.ts	Updated settings merge to use deep merge
packages/database/src/repositories/aiInfra/index.test.ts	Added tests for settings merge behavior
packages/database/src/models/aiModel.ts	Added settings field to model queries
locales/zh-CN/modelProvider.json	Added Chinese translations for extended parameters
locales/zh-CN/chat.json	Added Chinese translation for thinking budget

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

packages/model-runtime/src/providers/openrouter/index.ts

...ants]/(main)/settings/provider/features/ModelList/CreateNewModelModal/ExtendParamsSelect.tsx

packages/model-runtime/src/providers/vercelaigateway/index.ts

Copilot · 2026-01-04T06:45:04Z

packages/model-runtime/src/providers/openrouter/index.ts

+          if (hasReasoning && model.id.includes('gpt-5')) {
+            extendParams.push('gpt5ReasoningEffort');
+          }
+          if (hasReasoning && model.id.includes('openai') && !model.id.includes('gpt-5')) {
+            extendParams.push('reasoningEffort');
+          }
+          if (hasReasoning && model.id.includes('claude')) {
+            extendParams.push('enableReasoning', 'reasoningBudgetToken');
+          }
+          if (model.id.includes('claude') && writeCacheInputPrice && writeCacheInputPrice !== 0) {
+            extendParams.push('disableContextCaching');
+          }
+          if (hasReasoning && model.id.includes('gemini-2.5')) {


The same brittle string matching pattern exists in openrouter provider. Model ID checks using includes() are prone to false positives. For instance, model.id.includes('gpt-5') on line 129, model.id.includes('openai') on line 132, model.id.includes('claude') on lines 135 and 138, and model.id.includes('gemini-2.5') on line 141 can all match unintended model IDs. Consider implementing a more robust matching strategy with provider-specific prefixes or regex patterns.

Suggested change

if (hasReasoning && model.id.includes('gpt-5')) {

extendParams.push('gpt5ReasoningEffort');

}

if (hasReasoning && model.id.includes('openai') && !model.id.includes('gpt-5')) {

extendParams.push('reasoningEffort');

}

if (hasReasoning && model.id.includes('claude')) {

extendParams.push('enableReasoning', 'reasoningBudgetToken');

}

if (model.id.includes('claude') && writeCacheInputPrice && writeCacheInputPrice !== 0) {

extendParams.push('disableContextCaching');

}

if (hasReasoning && model.id.includes('gemini-2.5')) {

// More robust model id checks to avoid accidental matches on unrelated models

const isOpenAIGpt5Model =

hasReasoning && /^openai\/gpt-5([\/:\-]|$)/.test(model.id);

const isOpenAIReasoningModel =

hasReasoning &&

/^openai\//.test(model.id) &&

!/^openai\/gpt-5([\/:\-]|$)/.test(model.id);

const isClaudeModel = /(^|\/)claude([\/:\-]|$)/.test(model.id);

const isGemini25Model = /(^|\/)gemini-2\.5([\/:\-]|$)/.test(model.id);

if (isOpenAIGpt5Model) {

extendParams.push('gpt5ReasoningEffort');

}

if (isOpenAIReasoningModel) {

extendParams.push('reasoningEffort');

}

if (hasReasoning && isClaudeModel) {

extendParams.push('enableReasoning', 'reasoningBudgetToken');

}

if (isClaudeModel && writeCacheInputPrice && writeCacheInputPrice !== 0) {

extendParams.push('disableContextCaching');

}

if (hasReasoning && isGemini25Model) {

arvinxx · 2026-01-11T06:30:30Z

@sxjeru There is a conflict that needs to be resolved.

This comment was translated by Claude.

Original Content

@sxjeru 有冲突,需要解一下

vercel · 2026-01-11T07:16:23Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Review	Updated (UTC)
lobehub	Ready	Preview, Comment	Jan 11, 2026 7:30am

lobehubbot · 2026-01-11T07:55:37Z

❤️ Great PR @sxjeru ❤️

The growth of project is inseparable from user feedback and contribution, thanks for your contribution! If you are interested in the lobehub developer community, please join our discord and then dm @arvinxx or @canisminor1990. They will invite you to our private developer channel where we discuss lobe-chat development and share AI newsletters from around the world.

## [Version 2.0.0-next.263](v2.0.0-next.262...v2.0.0-next.263) <sup>Released on **2026-01-11**</sup> #### 💄 Styles - **misc**: ModelParse support to get model settings params & Customize extendParams UI. <br/> <details> <summary><kbd>Improvements and Fixes</kbd></summary> #### Styles * **misc**: ModelParse support to get model settings params & Customize extendParams UI, closes [#11185](#11185) ([94e985a](94e985a)) </details> <div align="right"> [![](https://img.shields.io/badge/-BACK_TO_TOP-151515?style=flat-square)](#readme-top) </div>

lobehubbot · 2026-01-11T08:16:01Z

🎉 This PR is included in version 2.0.0-next.263 🎉

The release is available on:

Your semantic-release bot 📦🚀

…extendParams UI (lobehub#11185) * ✨ feat: 添加扩展参数支持，增强模型配置功能 * 支持更多扩展参数；支持 openrouter * style: 添加思考预算扩展参数支持，更新相关标签 * Refactor sliders to use a unified LevelSlider component for improved code reusability and maintainability. Updated GPT52ProReasoningEffortSlider, GPT52ReasoningEffortSlider, GPT5ReasoningEffortSlider, ImageAspectRatioSelect, ImageResolutionSlider, ReasoningEffortSlider, TextVerbositySlider, ThinkingLevel2Slider, ThinkingLevelSlider, and ThinkingSlider to implement controlled and uncontrolled modes. Enhanced aspect ratio selection and resolution sliders with better type safety and default values. * Refactor reasoning effort sliders and related components to use a factory function for better code reuse and maintainability - Created `createLevelSliderComponent` to handle both controlled and uncontrolled modes for sliders. - Updated `GPT5ReasoningEffortSlider`, `GPT51ReasoningEffortSlider`, `GPT52ProReasoningEffortSlider`, `GPT52ReasoningEffortSlider`, `ImageResolutionSlider`, `ImageAspectRatioSelect`, `ReasoningEffortSlider`, `TextVerbositySlider`, `ThinkingSlider`, `ThinkingLevelSlider`, and `ThinkingLevel2Slider` to utilize the new factory function. - Simplified the logic for handling controlled and uncontrolled states in sliders. - Added tests for the new slider factory to ensure correct behavior in both controlled and uncontrolled modes. * Reuses i18n keys via alias mapping for param titles * fix tests

## [Version 2.0.0-next.263](lobehub/lobehub@v2.0.0-next.262...v2.0.0-next.263) <sup>Released on **2026-01-11**</sup> #### 💄 Styles - **misc**: ModelParse support to get model settings params & Customize extendParams UI. <br/> <details> <summary><kbd>Improvements and Fixes</kbd></summary> #### Styles * **misc**: ModelParse support to get model settings params & Customize extendParams UI, closes [lobehub#11185](lobehub#11185) ([94e985a](lobehub@94e985a)) </details> <div align="right"> [![](https://img.shields.io/badge/-BACK_TO_TOP-151515?style=flat-square)](#readme-top) </div>

## [Version 1.151.0](v1.150.0...v1.151.0) <sup>Released on **2026-01-16**</sup> #### ♻ Code Refactoring - **agentGroup**: Simplify LobeChatGroupChatConfig schema. - **database**: Renamed to userMemories/index.ts. - **desktop**: Unify TITLE_BAR_HEIGHT constant to desktop-bridge. - **misc**: Improve Tools popover component structure and fix UI consistency, migrate Next.js navigation APIs to React Router for SPA, refactor page and notebook document usage, remove the old lobehub plugins, rename chat folder to agent, use fallbackData to prevent useActionSWR auto-fetch. #### ✨ Features - **community**: Recommendation for agent & Discover tab, recommended for home & added discover tab, support to report for agent & mcp plugin interaction for recommendation. - **conversation**: Use native context menu when selection is within current ChatItem. - **desktop**: Add desktop release service and API endpoint, add local update testing scripts and stable channel API version check, implement history navigation stack, improve macOS permission requests and Full Disk Access detection. - **electron**: Add custom titlebar for Electron windows. - **share**: Add topic sharing functionality. - **userMemories**: Support to assign for extra headers when invoking upstash workflows, support to use customized Qstash client with extra header for workflows. - **misc**: Add the agent cron job, generate `agent_cron_jobs` in agents, improve baseline alignment for tool items, improve group profile builder, improve group prompt context engine and fix group supervisor response issue, improve PageEditor header UX with DropdownMenu and i18n support, update model definitions and sync i18n, update the agent profiles tools check & agentbuilder tools & publish to market button, update the community user layout action button, update the cron job visiual way, update the cron patterns fields values. #### 🐛 Bug Fixes - **@upstash/qstash**: Should properly extract Error message from thrown one. - **api**: Fix the issue where custom AI Providers cannot use custom APIs. - **ChatInput**: Add missing MaxTokens setting to params popover. - **chat**: Reset activeTopicId when switching agent/group. - **controls**: Update checkbox toggle behavior and pass value to ParamControlWrapper. - **database**: Remove content validation limits for agent cron jobs. - **desktop**: Prevent duplicate CORS headers in response, return OFFICIAL_URL in cloud mode for remoteServerUrl selector, update macOS beta icon size for macOS 26. - **InputEditor**: Ensure lexical placeholder reactively updates on locale change. - **mcp**: Fix installation check hanging issue in desktop app. - **settings**: Add instant UI feedback for provider config switches. - **misc**: Click lobe ai topic trigger create new agent, fix auto add group member crash, fix duplicate agent and group, fix group ux and memory retriever, fix internal editor onTextChange issue and add test case, fix memory search context, fix new topic flick issue, fix page content mismatch when switch quickly, fix thread portal not open correctly, Fix Windows desktop build error with macOS native module, force plain text paste in ChatInput editor, handle will-prevent-unload event to allow app quit, prevent auto navigation to profile when clicking topic, refresh sidebar after sendAsGroup and add E2E tests, slove the settings/profile change but not refresh the profiles. #### 💄 Styles - **desktop**: Update macOS beta icon assets. - **misc**: Add MiniMax-M2.1 and GLM-4.7 for Qiniu provider, fix UI issues with tooltip wrapping and dropdown type, force gpt-5.2 use response api, improve agent loading state, improve cron jobs style, improve todo list, modelParse support to get model settings params & Customize extendParams UI, remember page agent panel width, update i18n, update i18n, update readFile content. <br/> <details> <summary><kbd>Improvements and Fixes</kbd></summary> #### Code refactoring * **agentGroup**: Simplify LobeChatGroupChatConfig schema, closes [lobehub#11431](https://github.com/jaworldwideorg/OneJA-Bot/issues/11431) ([a8c6b2c](a8c6b2c)) * **database**: Renamed to userMemories/index.ts, closes [lobehub#11359](https://github.com/jaworldwideorg/OneJA-Bot/issues/11359) ([9dae5ff](9dae5ff)) * **desktop**: Unify TITLE_BAR_HEIGHT constant to desktop-bridge, closes [lobehub#11496](https://github.com/jaworldwideorg/OneJA-Bot/issues/11496) ([e7739e5](e7739e5)) * **misc**: Improve Tools popover component structure and fix UI consistency, closes [lobehub#11356](https://github.com/jaworldwideorg/OneJA-Bot/issues/11356) ([f46837a](f46837a)) * **misc**: Migrate Next.js navigation APIs to React Router for SPA, closes [lobehub#11394](https://github.com/jaworldwideorg/OneJA-Bot/issues/11394) ([2253d46](2253d46)) * **misc**: Refactor page and notebook document usage, closes [lobehub#11345](https://github.com/jaworldwideorg/OneJA-Bot/issues/11345) ([88721eb](88721eb)) * **misc**: Remove the old lobehub plugins, closes [lobehub#11498](https://github.com/jaworldwideorg/OneJA-Bot/issues/11498) ([e5b47df](e5b47df)) * **misc**: Rename chat folder to agent, closes [lobehub#11409](https://github.com/jaworldwideorg/OneJA-Bot/issues/11409) ([7cfb1a3](7cfb1a3)) * **misc**: Use fallbackData to prevent useActionSWR auto-fetch, closes [lobehub#11514](https://github.com/jaworldwideorg/OneJA-Bot/issues/11514) ([d446163](d446163)) #### What's improved * **community**: Recommendation for agent & Discover tab, closes [lobehub#11445](https://github.com/jaworldwideorg/OneJA-Bot/issues/11445) ([5c102b5](5c102b5)) * **community**: Recommended for home & added discover tab, closes [lobehub#11290](https://github.com/jaworldwideorg/OneJA-Bot/issues/11290) ([8db248c](8db248c)) * **community**: Support to report for agent & mcp plugin interaction for recommendation, closes [lobehub#11289](https://github.com/jaworldwideorg/OneJA-Bot/issues/11289) ([6f98792](6f98792)) * **conversation**: Use native context menu when selection is within current ChatItem, closes [lobehub#11400](https://github.com/jaworldwideorg/OneJA-Bot/issues/11400) ([9778dce](9778dce)) * **desktop**: Add desktop release service and API endpoint, closes [lobehub#11520](https://github.com/jaworldwideorg/OneJA-Bot/issues/11520) ([e3dc5be](e3dc5be)) * **desktop**: Add local update testing scripts and stable channel API version check, closes [lobehub#11474](https://github.com/jaworldwideorg/OneJA-Bot/issues/11474) [lobehub#11513](https://github.com/jaworldwideorg/OneJA-Bot/issues/11513) ([959c210](959c210)) * **desktop**: Implement history navigation stack, closes [lobehub#11341](https://github.com/jaworldwideorg/OneJA-Bot/issues/11341) ([db270d5](db270d5)) * **desktop**: Improve macOS permission requests and Full Disk Access detection, closes [lobehub#11380](https://github.com/jaworldwideorg/OneJA-Bot/issues/11380) ([2d5868f](2d5868f)) * **electron**: Add custom titlebar for Electron windows, closes [lobehub#11438](https://github.com/jaworldwideorg/OneJA-Bot/issues/11438) ([08f6ee3](08f6ee3)) * **share**: Add topic sharing functionality, closes [lobehub#11448](https://github.com/jaworldwideorg/OneJA-Bot/issues/11448) ([ddca165](ddca165)) * **userMemories**: Support to assign for extra headers when invoking upstash workflows, closes [lobehub#11374](https://github.com/jaworldwideorg/OneJA-Bot/issues/11374) ([895e15e](895e15e)) * **userMemories**: Support to use customized Qstash client with extra header for workflows, closes [lobehub#11378](https://github.com/jaworldwideorg/OneJA-Bot/issues/11378) ([3417af4](3417af4)) * **misc**: Add the agent cron job, closes [lobehub#11370](https://github.com/jaworldwideorg/OneJA-Bot/issues/11370) ([10e47d9](10e47d9)) * **misc**: Generate `agent_cron_jobs` in agents, closes [lobehub#11349](https://github.com/jaworldwideorg/OneJA-Bot/issues/11349) ([eefb6cb](eefb6cb)) * **misc**: Improve baseline alignment for tool items, closes [lobehub#11447](https://github.com/jaworldwideorg/OneJA-Bot/issues/11447) ([be8dddd](be8dddd)) * **misc**: Improve group profile builder, closes [lobehub#11452](https://github.com/jaworldwideorg/OneJA-Bot/issues/11452) ([9012b40](9012b40)) * **misc**: Improve group prompt context engine and fix group supervisor response issue, closes [lobehub#11490](https://github.com/jaworldwideorg/OneJA-Bot/issues/11490) ([7d066eb](7d066eb)) * **misc**: Improve PageEditor header UX with DropdownMenu and i18n support, closes [lobehub#11462](https://github.com/jaworldwideorg/OneJA-Bot/issues/11462) ([ae499c9](ae499c9)) * **misc**: Update model definitions and sync i18n, closes [lobehub#11468](https://github.com/jaworldwideorg/OneJA-Bot/issues/11468) ([484ffb3](484ffb3)) * **misc**: Update the agent profiles tools check & agentbuilder tools & publish to market button, closes [lobehub#11501](https://github.com/jaworldwideorg/OneJA-Bot/issues/11501) ([85277fa](85277fa)) * **misc**: Update the community user layout action button, closes [lobehub#11472](https://github.com/jaworldwideorg/OneJA-Bot/issues/11472) ([2dd6d42](2dd6d42)) * **misc**: Update the cron job visiual way, closes [lobehub#11466](https://github.com/jaworldwideorg/OneJA-Bot/issues/11466) ([63d81de](63d81de)) * **misc**: Update the cron patterns fields values, closes [lobehub#11399](https://github.com/jaworldwideorg/OneJA-Bot/issues/11399) ([7632cef](7632cef)) #### What's fixed * **@upstash/qstash**: Should properly extract Error message from thrown one, closes [lobehub#11465](https://github.com/jaworldwideorg/OneJA-Bot/issues/11465) ([a8142b4](a8142b4)) * **api**: Fix the issue where custom AI Providers cannot use custom APIs, closes [lobehub#11335](https://github.com/jaworldwideorg/OneJA-Bot/issues/11335) ([2c666b8](2c666b8)) * **ChatInput**: Add missing MaxTokens setting to params popover, closes [lobehub#11412](https://github.com/jaworldwideorg/OneJA-Bot/issues/11412) [lobehub#11375](https://github.com/jaworldwideorg/OneJA-Bot/issues/11375) ([3db4389](3db4389)) * **chat**: Reset activeTopicId when switching agent/group, closes [lobehub#11523](https://github.com/jaworldwideorg/OneJA-Bot/issues/11523) ([fde54b0](fde54b0)) * **controls**: Update checkbox toggle behavior and pass value to ParamControlWrapper, closes [lobehub#11363](https://github.com/jaworldwideorg/OneJA-Bot/issues/11363) ([1f1ef94](1f1ef94)) * **database**: Remove content validation limits for agent cron jobs, closes [lobehub#11444](https://github.com/jaworldwideorg/OneJA-Bot/issues/11444) ([04a28d3](04a28d3)) * **desktop**: Prevent duplicate CORS headers in response, closes [lobehub#11350](https://github.com/jaworldwideorg/OneJA-Bot/issues/11350) ([57e725c](57e725c)) * **desktop**: Return OFFICIAL_URL in cloud mode for remoteServerUrl selector, closes [lobehub#11502](https://github.com/jaworldwideorg/OneJA-Bot/issues/11502) ([1d11fac](1d11fac)) * **desktop**: Update macOS beta icon size for macOS 26, closes [lobehub#11348](https://github.com/jaworldwideorg/OneJA-Bot/issues/11348) ([0d1eedf](0d1eedf)) * **InputEditor**: Ensure lexical placeholder reactively updates on locale change, closes [lobehub#11352](https://github.com/jaworldwideorg/OneJA-Bot/issues/11352) ([72e796b](72e796b)) * **mcp**: Fix installation check hanging issue in desktop app, closes [lobehub#11524](https://github.com/jaworldwideorg/OneJA-Bot/issues/11524) ([b9341c3](b9341c3)) * **settings**: Add instant UI feedback for provider config switches, closes [lobehub#11362](https://github.com/jaworldwideorg/OneJA-Bot/issues/11362) ([a758d01](a758d01)) * **misc**: Click lobe ai topic trigger create new agent, closes [lobehub#11508](https://github.com/jaworldwideorg/OneJA-Bot/issues/11508) ([2443189](2443189)) * **misc**: Fix auto add group member crash, closes [lobehub#11387](https://github.com/jaworldwideorg/OneJA-Bot/issues/11387) ([fe4ff91](fe4ff91)) * **misc**: Fix duplicate agent and group, closes [lobehub#11411](https://github.com/jaworldwideorg/OneJA-Bot/issues/11411) ([bc8aea4](bc8aea4)) * **misc**: Fix group ux and memory retriever, closes [lobehub#11481](https://github.com/jaworldwideorg/OneJA-Bot/issues/11481) ([033ca92](033ca92)) * **misc**: Fix internal editor onTextChange issue and add test case, closes [lobehub#11509](https://github.com/jaworldwideorg/OneJA-Bot/issues/11509) ([e5eb03e](e5eb03e)) * **misc**: Fix memory search context, closes [lobehub#11393](https://github.com/jaworldwideorg/OneJA-Bot/issues/11393) ([9f51a4c](9f51a4c)) * **misc**: Fix new topic flick issue, closes [lobehub#11473](https://github.com/jaworldwideorg/OneJA-Bot/issues/11473) ([c53d372](c53d372)) * **misc**: Fix page content mismatch when switch quickly, closes [lobehub#11505](https://github.com/jaworldwideorg/OneJA-Bot/issues/11505) ([0cb1374](0cb1374)) * **misc**: Fix thread portal not open correctly, closes [lobehub#11475](https://github.com/jaworldwideorg/OneJA-Bot/issues/11475) ([e6ff90b](e6ff90b)) * **misc**: Fix Windows desktop build error with macOS native module, closes [lobehub#11417](https://github.com/jaworldwideorg/OneJA-Bot/issues/11417) ([67a8114](67a8114)) * **misc**: Force plain text paste in ChatInput editor, closes [lobehub#11414](https://github.com/jaworldwideorg/OneJA-Bot/issues/11414) ([70daf13](70daf13)) * **misc**: Handle will-prevent-unload event to allow app quit, closes [lobehub#11406](https://github.com/jaworldwideorg/OneJA-Bot/issues/11406) ([cbeb013](cbeb013)) * **misc**: Prevent auto navigation to profile when clicking topic, closes [lobehub#11500](https://github.com/jaworldwideorg/OneJA-Bot/issues/11500) ([1e03005](1e03005)) * **misc**: Refresh sidebar after sendAsGroup and add E2E tests, closes [lobehub#11450](https://github.com/jaworldwideorg/OneJA-Bot/issues/11450) ([8376a80](8376a80)) * **misc**: Slove the settings/profile change but not refresh the profiles, closes [lobehub#11497](https://github.com/jaworldwideorg/OneJA-Bot/issues/11497) ([f1e2111](f1e2111)) #### Styles * **desktop**: Update macOS beta icon assets, closes [lobehub#11368](https://github.com/jaworldwideorg/OneJA-Bot/issues/11368) ([3623e58](3623e58)) * **misc**: Add MiniMax-M2.1 and GLM-4.7 for Qiniu provider, closes [lobehub#10982](https://github.com/jaworldwideorg/OneJA-Bot/issues/10982) ([695784d](695784d)) * **misc**: Fix UI issues with tooltip wrapping and dropdown type, closes [lobehub#11495](https://github.com/jaworldwideorg/OneJA-Bot/issues/11495) ([9d90eba](9d90eba)) * **misc**: Force gpt-5.2 use response api, closes [lobehub#11373](https://github.com/jaworldwideorg/OneJA-Bot/issues/11373) ([f7f3631](f7f3631)) * **misc**: Improve agent loading state, closes [lobehub#11511](https://github.com/jaworldwideorg/OneJA-Bot/issues/11511) ([3bb7f33](3bb7f33)) * **misc**: Improve cron jobs style, closes [lobehub#11420](https://github.com/jaworldwideorg/OneJA-Bot/issues/11420) ([d1602f6](d1602f6)) * **misc**: Improve todo list, closes [lobehub#11533](https://github.com/jaworldwideorg/OneJA-Bot/issues/11533) ([a4b71e9](a4b71e9)) * **misc**: ModelParse support to get model settings params & Customize extendParams UI, closes [lobehub#11185](https://github.com/jaworldwideorg/OneJA-Bot/issues/11185) ([94e985a](94e985a)) * **misc**: Remember page agent panel width, closes [lobehub#11389](https://github.com/jaworldwideorg/OneJA-Bot/issues/11389) ([801b624](801b624)) * **misc**: Update i18n, closes [lobehub#11425](https://github.com/jaworldwideorg/OneJA-Bot/issues/11425) ([87fbed2](87fbed2)) * **misc**: Update i18n, closes [lobehub#11360](https://github.com/jaworldwideorg/OneJA-Bot/issues/11360) ([da09825](da09825)) * **misc**: Update readFile content, closes [lobehub#11485](https://github.com/jaworldwideorg/OneJA-Bot/issues/11485) ([050499b](050499b)) </details> <div align="right"> [![](https://img.shields.io/badge/-BACK_TO_TOP-151515?style=flat-square)](#readme-top) </div>

sxjeru and others added 6 commits January 3, 2026 22:50

✨ feat: 添加扩展参数支持，增强模型配置功能

d3d9010

支持更多扩展参数；支持 openrouter

b6481e5

style: 添加思考预算扩展参数支持，更新相关标签

8088462

Merge branch 'lobehub:next' into extendparam2

7140a46

sourcery-ai bot reviewed Jan 4, 2026

View reviewed changes

sxjeru added 2 commits January 4, 2026 14:12

Reuses i18n keys via alias mapping for param titles

af9ba7f

fix tests

2701de2

sxjeru marked this pull request as ready for review January 4, 2026 06:40

Copilot AI review requested due to automatic review settings January 4, 2026 06:40

dosubot bot added size:XXL This PR changes 1000+ lines, ignoring generated files. 💄 Design Design and style issues labels Jan 4, 2026

Copilot started reviewing on behalf of sxjeru January 4, 2026 06:41 View session

sourcery-ai bot reviewed Jan 4, 2026

View reviewed changes

Copilot AI reviewed Jan 4, 2026

View reviewed changes

Merge branch 'next' into extendparam2

5877cd5

vercel bot deployed to Preview January 11, 2026 07:30 View deployment

arvinxx merged commit 94e985a into lobehub:next Jan 11, 2026
24 checks passed

sxjeru deleted the extendparam2 branch January 11, 2026 08:05

lobehubbot added the released on @next label Jan 11, 2026

dosubot bot mentioned this pull request Jan 11, 2026

Model parameter description displays behind the adjustment panel #11419

Closed

8 tasks

heidenhm mentioned this pull request Jan 11, 2026

Custom extend parameters don't seem to take effect #11424

Closed

8 tasks

Uh oh!

💄 style: modelParse support to get model settings params & Customize extendParams UI #11185

💄 style: modelParse support to get model settings params & Customize extendParams UI #11185

Uh oh!

Conversation

sxjeru commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💻 Change Type

🔗 Related Issue

🔀 Description of Change

🧪 How to Test

📸 Screenshots / Videos

📝 Additional Information

Uh oh!

vercel bot commented Jan 4, 2026

Uh oh!

sourcery-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

gru-agent bot commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sourcery-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI Jan 4, 2026

Choose a reason for hiding this comment

Uh oh!

arvinxx commented Jan 11, 2026 • edited by lobehubbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vercel bot commented Jan 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

lobehubbot commented Jan 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lobehubbot commented Jan 11, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

sxjeru commented Jan 4, 2026 •

edited

Loading

codecov bot commented Jan 4, 2026 •

edited

Loading

gru-agent bot commented Jan 4, 2026 •

edited

Loading

arvinxx commented Jan 11, 2026 •

edited by lobehubbot

Loading

vercel bot commented Jan 11, 2026 •

edited

Loading

lobehubbot commented Jan 11, 2026 •

edited

Loading