ghostplant
Results
272
comments of
ghostplant
But it is a good feature request and we'll handle those settings through REST API in the future.
The new version of `tutelgroup/deepseek-671b:mi300x8-chat-20250723` allows setting `max_tokens` in REST: ```sh curl -X POST http://0.0.0.0:8000/chat -d '{"messages": [{"role": "user", "content": "Hello."}], "max_tokens": 256 }' ```