ImportError
#26
by
kris-123
- opened
Following is the error trying to run zephyr-7b-alpha using HuggingFaceLLM
method in llama_index.llms
locally:
ImportError: Using load_in_8bit=True
requires Accelerate: pip install accelerate
and the latest version of bitsandbytes pip install -i https://test.pypi.org/simple/ bitsandbytes
or pip install bitsandbytes
Libraries installed already:
transformers - tried both 4.34.1 and 4.35.0.dev0
accelerate - 0.23.0
bistandbytes - 0.41.1
+1, seems that whatever I install in the notebook is not visible from within the huggingface lib that llama index uses