Spaces:

open-llm-leaderboard

/

open_llm_leaderboard

like 11.7k

Running on CPU Upgrade

I tried to add OpenAssistant/falcon-40b-sft-mix-1226, but it fails with the error Model "OpenAssistant/falcon-40b-sft-mix-1226"was not found on hub! even though it is available. I had a look at the app.py of this huggingface space and I believe the reason is that the code checks whether the model is available by calling AutoConfig.from_pretrained(model_name, revision=revision) in the is_model_on_hub function. Trying this locally does indeed fail for OpenAssistant/falcon-40b-sft-mix-1226 because the falcon based model requires trust_remote_code=True. Setting this then makes it work. Would it be possible to fix this?