Hugging Face's logo

transformers-community
/
contrastive-search

Text Generation
Transformers
Safetensors
qwen3
custom_generate
conversational
text-generation-inference
Model card Files
xet
Community
4

Fix cache format

#2
by cyrilvallez HF Staff - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+60
-177
cyrilvallez
•
edited Oct 9, 2025

Transformers now only uses the Cache classes so we cannot index past kv

 Fix cache formata4fcf5b1
 Update custom_generate/generate.py30636bc5
 Update custom_generate/generate.pyb9006ffa
 Update custom_generate/generate.py6cfd5a40
cyrilvallez changed pull request status to closed

Sign up or log in to comment