Spaces:
Paused
Paused
Commit History
-
"fix_deep_stream_interrupted_and_timeout"
378caa6
-
"fix_retry_timeout_for_cloudflare_524"
25bce68 -
"fix_backend_retry_logic"
dbb1ef4 -
"fix_frontend_image_compression"
81934c5
-
fix: 修复 /auth/config.js 端点缺少 g_config 导入的问题
9de587b -
fix: 恢复前端 API 密钥端点供内置 Web UI 使用
b759462 -
feat: 实现反检测速率限制器，模拟人类行为模式避免被 Google 封禁
4e293c6 -
feat: 实现智能速率限制器，支持自适应超时、队列管理和监控指标
6b2240a -
fix: 修复速率限制器死锁风险、临时文件清理兼容性和图片目录初始化问题
d4a9626 -
fix(P0/P1/P2): 添加 Cookie 持久化、临时文件清理、速率限制和移除 API 密钥暴露
de8b836
-
update: 同步最新代码与配置
b2c2a89
-
update: 更新聊天服务、客户端连接池和辅助工具
ac10cac
-
update: 更新前端样式、模型、聊天服务和客户端代码
f234f8e
-
修复流式输出错乱问题
b93f165
-
终极代码
0c0c606 -
feat: integrate frontend with health monitoring and align with upstream
ebd8962 -
:bug: Update function call format to avoid potential leak issues (#96)
Lưu Quang VũUnverified signature* Remove the unused auto-refresh functionality and related imports. They are no longer needed since the underlying library issue has been resolved. * Enhance error handling in client initialization and message sending * Refactor link handling to extract file paths and simplify Google search links * Fix regex pattern for Google search link matching * Fix regex patterns for Markdown escaping, code fence and Google search link matching * Increase timeout value in configuration files from 60 to 120 seconds to better handle heavy tasks * Fix Image generation * Refactor tool handling to support standard and image generation tools separately * Fix: use "ascii" decoding for base64-encoded image data consistency * Fix: replace `running` with `_running` for internal client status checks * Refactor: replace direct `_running` access with `running()` method in client status checks * Extend models with new fields for annotations, reasoning, audio, log probabilities, and token details; ad…7fdfe87
-
:sparkles: Enable real-time streaming responses and completely solve the issue with reusable sessions. (#95)
Lưu Quang VũUnverified signature* Remove the unused auto-refresh functionality and related imports. They are no longer needed since the underlying library issue has been resolved. * Enhance error handling in client initialization and message sending * Refactor link handling to extract file paths and simplify Google search links * Fix regex pattern for Google search link matching * Fix regex patterns for Markdown escaping, code fence and Google search link matching * Increase timeout value in configuration files from 60 to 120 seconds to better handle heavy tasks * Fix Image generation * Refactor tool handling to support standard and image generation tools separately * Fix: use "ascii" decoding for base64-encoded image data consistency * Fix: replace `running` with `_running` for internal client status checks * Refactor: replace direct `_running` access with `running()` method in client status checks * Extend models with new fields for annotations, reasoning, audio, log probabilities, and token details; ad…7db4283
-
refactor: Optimize performance by reducing token usage and speed up model response time. (#89)
Lưu Quang VũUnverified signature* Remove the unused auto-refresh functionality and related imports. They are no longer needed since the underlying library issue has been resolved. * Enhance error handling in client initialization and message sending * Refactor link handling to extract file paths and simplify Google search links * Fix regex pattern for Google search link matching * Fix regex patterns for Markdown escaping, code fence and Google search link matching * Increase timeout value in configuration files from 60 to 120 seconds to better handle heavy tasks * Fix Image generation * Refactor tool handling to support standard and image generation tools separately * Fix: use "ascii" decoding for base64-encoded image data consistency * Fix: replace `running` with `_running` for internal client status checks * Refactor: replace direct `_running` access with `running()` method in client status checks * Extend models with new fields for annotations, reasoning, audio, log probabilities, and token details; ad…f8272eb
-
feat: Add support for custom Gemini models and model loading strategies (#86)
Lưu Quang VũUnverified signature* feat: Add support for custom Gemini models and model loading strategies - Introduced `model_strategy` configuration for "append" (default + custom models) or "overwrite" (custom models only). - Enhanced `/v1/models` endpoint to return models based on the configured strategy. - Improved model loading with environment variable overrides and validation. - Refactored model handling logic for improved modularity and error handling. * feat: Improve Gemini model environment variable parsing and nested field support - Enhanced `extract_gemini_models_env` to handle nested fields within environment variables. - Updated type hints for more flexibility in model overrides. - Improved `_merge_models_with_env` to better support field-level updates and appending new models. * refactor: Consolidate utility functions and clean up unused code - Moved utility functions like `strip_code_fence`, `extract_tool_calls`, and `iter_stream_segments` to a centralized helper module. - Removed unused and redu…a583ded
-
:sparkles: Update the /v1/responses endpoint (#75)
Lưu Quang VũUnverified signature* Remove the unused auto-refresh functionality and related imports. They are no longer needed since the underlying library issue has been resolved. * Enhance error handling in client initialization and message sending * Refactor link handling to extract file paths and simplify Google search links * Fix regex pattern for Google search link matching * Fix regex patterns for Markdown escaping, code fence and Google search link matching * Increase timeout value in configuration files from 60 to 120 seconds to better handle heavy tasks * Fix Image generation * Refactor tool handling to support standard and image generation tools separately * Fix: use "ascii" decoding for base64-encoded image data consistency * Fix: replace `running` with `_running` for internal client status checks * Refactor: replace direct `_running` access with `running()` method in client status checks * Extend models with new fields for annotations, reasoning, audio, log probabilities, and token details; ad…5cb387e
-
:sparkles: Optimize Response API, Link Handling, and Add File Upload Support (#73)
Lưu Quang VũUnverified signature* Remove the unused auto-refresh functionality and related imports. They are no longer needed since the underlying library issue has been resolved. * Enhance error handling in client initialization and message sending * Refactor link handling to extract file paths and simplify Google search links * Fix regex pattern for Google search link matching * Fix regex patterns for Markdown escaping, code fence and Google search link matching * Increase timeout value in configuration files from 60 to 120 seconds to better handle heavy tasks * Fix Image generation * Refactor tool handling to support standard and image generation tools separately * Fix: use "ascii" decoding for base64-encoded image data consistency * Fix: replace `running` with `_running` for internal client status checksff54322
-
:sparkles: Add proxy support and fix bugs (#56)
Yongkun Li +1: Lưu Quang VũUnverified signature* Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add endpoint generate and edit images * Add endpoint generate and edit images * Add support structured output * Add support structured output * Incorrect logging * Force LLM to follow tool_call format * Format the code using Black. * Fixes "Error handling message: No image returned" * Return image dimensions * Fixes Pylance warning * Format by ruff * uv run directly * Fixes XML_WRAP_HINT leaked * Instruct an LLM to return code snippets enclosed within Markdown fenced code blocks. * Adjust the streaming response logic to ensure important sections remain intact and are not fragmented. * Change chunk_size to 64 * ruff check * Fixes for im_start/im_end hints leaking from responses. * Ensure all endpoints are fully compliant with OpenAI compatibility standards. * :memo: Fix doc * Add a retention_days option to schedule database cleanup. - F…ce9e9da
-
:sparkles: Fix bugs and add an option to schedule database cleanup. (#49)
Lưu Quang Vũ +1: Nativu5Unverified signature* Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add endpoint generate and edit images * Add endpoint generate and edit images * Add support structured output * Add support structured output * Incorrect logging * Force LLM to follow tool_call format * Format the code using Black. * Fixes "Error handling message: No image returned" * Return image dimensions * Fixes Pylance warning * Format by ruff * uv run directly * Fixes XML_WRAP_HINT leaked * Instruct an LLM to return code snippets enclosed within Markdown fenced code blocks. * Adjust the streaming response logic to ensure important sections remain intact and are not fragmented. * Change chunk_size to 64 * ruff check * Fixes for im_start/im_end hints leaking from responses. * Ensure all endpoints are fully compliant with OpenAI compatibility standards. * :memo: Fix doc * Add a retention_days option to schedule database cleanup. - F…68b21f6
-
:sparkles: Add tool call support and fix image input and output (#47)
Lưu Quang Vũ +1: Nativu5Unverified signature* Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add tool_calls support * Add endpoint generate and edit images * Add endpoint generate and edit images * Add support structured output * Add support structured output * Incorrect logging * Force LLM to follow tool_call format * Format the code using Black. * Fixes "Error handling message: No image returned" * Return image dimensions * Fixes Pylance warning * Format by ruff * uv run directly * Fixes XML_WRAP_HINT leaked * Instruct an LLM to return code snippets enclosed within Markdown fenced code blocks. * Adjust the streaming response logic to ensure important sections remain intact and are not fragmented. * Change chunk_size to 64 * ruff check * Fixes for im_start/im_end hints leaking from responses. * Ensure all endpoints are fully compliant with OpenAI compatibility standards. * :memo: Fix doc --------- Co-authored-by: Nativu5 <44155313+Nativu5@users.norep…77830e0
-
:sparkles: Add HTTPS mode (#39)
Yongkun Li +2: MinQ9362, Copilot* feat: add https for fastapi * fix: fix * Update README.md * fix: fix import os * fix: add logger info * Update run.py * Update pyproject.toml * fix: Update run.py * feat: Add HTTPS configuration support and update gemini-webapi dependency * fix: docs * fix: typos Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> * fix: Fix comments --------- Co-authored-by: MinQ <shenmintao@gmail.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>be6825e
-
:bug: Fix XML wrap hint formatting in client.py (#27)
David ZhangUnverified signature* :bug: Fix XML wrap hint formatting in client.py2a32850
-
:sparkles: Add retrying to fix ModelInvalid error (#23)
Yongkun LiUnverified signature* feat: Add Ruff in dev dependency * fix: Add retrying for ModelInvalid error * fix: Add re-raisecba6b3c
-
:bug: Fix tagging when using XML (#18)
Yongkun Li +1: CopilotUnverified signature* :art: Fix lint error * Fix XML hint logic (#17) * :bug: Fix bugs * :bug: Update with regex Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>bea78d0
-
:sparkles: Multi-segment text merging, ultra-long message splitting, xml prompt, tag, session, and token calculation optimization. (#16)
David ZhangUnverified signature* :sparkles: Enhance text input handling by appending multiple fragments * :bulb: Add guidance for wrapping XML blocks in responses * :bulb: Skip tagging for the first message in a new session. * :sparkles: Implement message splitting for Gemini requests exceeding the character limit * :sparkles: try to reuse session as much as possible * :bulb: fix response token calculation to include the whole history * :bulb: Enhance streaming response to include token usage calculation7f67a14
-
:recycle: Refactor history sanitizing (#12)
Yongkun LiUnverified signatureb2f35b6
-
:bug: Fix session persistence, when request contains <think> (#10)
David ZhangUnverified signature3481315 -
:sparkles: Implement Gemini client pool (#7)
Yongkun Li +1: CopilotUnverified signature* 📝 docs: clarify client id usage * :sparkles: Add logging and checking * 🔧 fix env override for list items * :bug: Fix config reading and singleton of client pool * :loud_sound: Add more logs Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>93d79eb
-
:art: Fix format and update docs (#6)
Yongkun LiUnverified signature* :bug: Fix <think> format * :art: Update Dockerfile and README files for clarity9d88248 -
:bug: Fix LMDB history mismatch (#4)
Yongkun LiUnverified signaturea487b1e
-
:pencil2: Fix typo in method name (#3)
Yongkun LiUnverified signatureab0396f -
:bug: Fix lmdb health check
Nativu570ff909 -
:sparkles: Add hash for model name and fix bugs
Nativu52ad2310 -
:bug: Fix broken multiple line streaming
Nativu5e17d521 -
:pencil: Fix docs
Nativu56c09e56 -
:sparkles: Add support for multi-round conversation
Nativu5cd70b5d
-
:bug: Fix bugs and optimize file uploading
Nativu5f2fc66c -
:construction: Refactored and fix bugs
Nativu54f6a1a5
-
:construction: Refactored with new structure
Nativu5f3518c5