fix: media timeouts + Qwen-Image-2.1 fixes + unified Team chat search - #327
Merged
Merged
Conversation
…trl+F on macOS - Keyboard: Team chat find only triggers on Cmd+F on macOS (Ctrl+F falls through to the native cursor-forward default); Win/Linux keep Ctrl+F. Matches the isMac pattern used in App.tsx. - Merge the two search UIs: find-in-conversation now ALSO queries the server FTS5 index (debounced) and lists full-history results below the instant local matches. Removes the separate ChatSearchPanel entry point, its header button, and the duplicate affordance. - ChatHistorySearch gains optional serverResults/serverLoading/ onServerResultClick props + 'All history' section (en/zh-CN/es).
P0-1 模型能力声明缺失 → 本地/自托管模型无法路由到多模态能力： - llm_add_model 新增 capabilities 参数（imageGeneration/vision/tts/stt/videoGeneration/decision）， 显式声明优先于模型 id 命名启发式，并持久化到 customModels - llm_set_capability_routing 新增 force 开关：绕过命名启发式（仍尊重 catalog 显式声明） - 校验提示给出三条出路：选可用模型 / llm_add_model 声明能力 / force 强制路由 - llm_list_providers 输出 capabilities 元数据；image_generation 命名模式扩展本地模型家族 （qwen-image/hunyuan/stable-diffusion/nano-banana 等） P0-2 硬编码媒体超时（120s 固定）→ 慢速本地推理被误杀： - LLMProviderConfig 新增 imageGenerationTimeoutMs/ttsTimeoutMs/sttTimeoutMs/ videoGenerationTimeoutMs/decisionTimeoutMs，OpenAI/MiniMax/DashScope/Fireworks/ Markus Provider 全部接入（默认值与历史硬编码一致，可覆盖） - llm_add_provider/llm_edit_provider 支持 *_timeout_ms 参数并持久化 P1-3 多模态 fallback 携带文本模型 id → 图像接口 404： - router.resolveModalityProvider 非文本能力不再复用全局文本路由默认模型 id， 仅当目录声明该模型具备对应能力时才携带 model - multimodal 候选解析：限定 provider/model id（如 deepseek/deepseek-v4-flash-0731） 拆分为显式 provider 走能力校验，报"不支持该模态"而非裸打图像接口 - generate_image 失败提示点明"文本模型不是图像模型，查 usable_models" P1-4 usable_models 与路由校验能力判定口径统一： - 单一查找入口 findModelDeclaredCapabilities 供校验与展示共用 测试：新增 capabilities 声明优先/force/媒体超时/qualified-id 拆分/路由模型能力判定 共 5 个测试文件 174+ 用例，全部通过。
OpenAI/MiniMax/DashScope/Fireworks/Markus 五个 Provider 的图像生成、TTS、 STT、视频生成四类媒体请求默认超时从 120s/180s 统一提高到 600s（10 分钟）， 覆盖本地/自托管推理服务器（diffusers、vLLM 等）首次请求需冷加载权重的场景。 decide（决策模型）非多模态，保持 60s 默认不变。 - LLMProviderConfig 注释同步；llm_add_provider/llm_edit_provider 工具描述更新 - 新增 OpenAIProvider/MarkusProvider 默认值断言测试
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Three commits on top of v0.10.0, targeting the 0.10.1-rc.0 line:
Test