Command -> litgpt download openlm-research/open_llama_13b gives error: Unrecognized arguments: openlm-research/open_llama_13b about litgpt HOT 4 OPEN

We recently updated the CLI, and it looks like the LitGPT version you are using requires the old --checkpoint_dir. I can make a new version release today for PyPI to address that. In the meantime, please upgrade LitGPT by

pip install litgpt==0.4.0.dev0

Let me know if it works or doesn't work.

from litgpt.

Yeah it is working , One more issue is when running this -> i# 3) Use the server (in a separate session) import requests, json response = requests.post( "http://127.0.0.1:8000/predict", json={"prompt": "Why people don't check cctv footage of all poling booths in INDIA to check if any malpractices occured? Is this not a big threat to democracy? "} ) print(response.json()) am getting output as "{'detail': 'Request timed out'}"

from litgpt.

Thanks for the feedback. Could you try a smaller model to test it with. Maybe the model is too large for your hardware and that's why it could be slow. E.g,. can you try

litgpt download EleutherAI/pythia-160m
litgpt serve EleutherAI/pythia-160m

from litgpt.