issues
search
redotvideo
/
haven
LLM fine-tuning and eval
https://haven.run
Apache License 2.0
341
stars
11
forks
source link
Add stop tokens to complete call, enable complete-only model-workers
#53
Closed
hkonsti
closed
1 year ago
hkonsti
commented
1 year ago
todo:
check that the error thrown inside the chat-completion handler on the worker is propagated to the sdk
todo: