10 Commits

Author SHA1 Message Date
yangdx
2985d88f97 refactor: improve CORS and streaming response headers
- Add configurable CORS origins
- Remove duplicate CORS headers
- Add X-Accel-Buffering header
- Update env example file
- Clean up header configurations
2025-02-15 11:39:10 +08:00
yangdx
28b17b327b Fix: top_k param handling error, unify top_k and cosine default value. 2025-02-13 06:05:21 +08:00
ArnoChen
88d691deb9 add namespace prefix to storage namespaces 2025-02-08 13:53:00 +08:00
yangdx
9103e7f463 fix: improve timing accuracy and variable scoping in OllamaAPI 2025-02-06 10:42:49 +08:00
yangdx
e124ad7f9c Fix timing calculation logic in OllamaAPI stream generators
• Initialize first_chunk_time as None
• Set timing only when first chunk arrives
2025-02-06 04:53:05 +08:00
yangdx
65dc0a6cfd Fix linting 2025-02-06 02:50:27 +08:00
yangdx
e26c6e564d refactor: enhance stream error handling and optimize code structure
- Initialize timestamps at start to avoid null checks
- Add detailed error handling for streaming response
- Handle CancelledError and other exceptions separately
- Unify exception handling with trace_exception
- Clean up redundant code and simplify logic
2025-02-06 02:43:06 +08:00
yangdx
db9b4dc841 Added environment variable loading with dotenv in Ollama API 2025-02-06 01:00:49 +08:00
yangdx
1a61d9ee7f Fix linting 2025-02-05 22:29:07 +08:00
yangdx
f703334ce4 Split the Ollama API implementation to a separated file 2025-02-05 22:15:14 +08:00