Spaces:
Sleeping
Sleeping
Commit History
-
Update waiting state messaging in app.py to provide clearer feedback when the endpoint is inactive. Adjust tests to validate the new placeholder behavior and ensure proper spinner functionality during waiting periods.
b3b0bdf -
Refactor endpoint handling in app.py to utilize a NamedTuple for model endpoints, improving token management and logging. Update README to reflect new environment variable requirements and enhance test coverage for model-specific token usage and error handling.
05ee7f4 -
Update app.py to increase MAX_TOKENS for reasoning models, introduce THINK_OPEN and THINK_CLOSE tags for reasoning content, and enhance chat controls. Add new chat functionality and improve model labeling. Update tests to validate reasoning handling and chat controls visibility.
9b4a365 -
Refactor model handling in chat app to support multiple endpoints for Gemma, Qwen, and DeepSeek. Update README with new environment variable requirements and enhance tests for model-specific endpoint functionality.
28f63c1
-
Fix UTF-8 decoding for SSE responses to prevent mojibake in non-ASCII text
9781ea9
-
Add support for cold endpoint handling with retry logic and timeout configuration
cba7f69 -
Add logging for streaming responses and enhance README configuration details
a8723cc
-
Enhance chat app functionality with incremental streaming and control locking
9787d64 -
Expose demo for Gradio Spaces hot reload and update main launch logic
b79d615 -
Implement streaming responses in chat app and update tests for new behavior
f072d11 -
Add .gitignore, update requirements, and implement tests for chat app functionality
1637033