Spaces:

---

X-Pipe

/

flash

like 0

Sleeping

[bugfix]fix the cut-off issue due to LLM predict token limit(256 for openai python lib default), by setting temperature to 0 and set LLM predict method from compact-refine to refine