Really challenging to do it in a space sdk
#35
by
deleted
- opened
You may want to try the quantized version of this model which will have lower memory requirements -- https://huggingface.co/CohereForAI/c4ai-command-r-v01-4bit.
sarahooker
changed discussion status to
closed
Btw, we now have launched a space https://huggingface.co/spaces/CohereForAI/c4ai-command-r-v01 which should allow you to play with the model.