No description provided.

to fix the load model issue:

$ python -m scripts.chat_web
2025-10-14 23:03:35,905 - nanochat.common - INFO - Distributed world size: 1
Starting NanoChat Web Server
Temperature: 0.8, Top-k: 50, Max tokens: 512
INFO:     Started server process [14892]
INFO:     Waiting for application startup.
Loading nanochat model...
ERROR:    Traceback (most recent call last):
  File "/home/ubuntu/karpathy/nanochat/.venv/lib/python3.13/site-packages/starlette/routing.py", line 694, in lifespan
    async with self.lifespan_context(app) as maybe_state:
               ~~~~~~~~~~~~~~~~~~~~~^^^^^
  File "/home/ubuntu/.local/share/uv/python/cpython-3.13.8-linux-x86_64-gnu/lib/python3.13/contextlib.py", line 214, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/karpathy/nanochat/scripts/chat_web.py", line 52, in lifespan
    app.state.model, app.state.tokenizer, _ = load_model(args.source, device, phase="eval", model_tag=args.model_tag, step=args.step)
                                              ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/karpathy/nanochat/nanochat/checkpoint_manager.py", line 146, in load_model
    return load_model_from_dir(checkpoints_dir, *args, **kwargs)
  File "/home/ubuntu/karpathy/nanochat/nanochat/checkpoint_manager.py", line 125, in load_model_from_dir
    model_tag = find_largest_model(checkpoints_dir)
  File "/home/ubuntu/karpathy/nanochat/nanochat/checkpoint_manager.py", line 95, in find_largest_model
    raise FileNotFoundError(f"No checkpoints found in {checkpoint_dir}")
FileNotFoundError: No checkpoints found in /home/ubuntu/.cache/nanochat/chatsft_checkpoints

ERROR:    Application startup failed. Exiting.

Thanks very much for the PR! Merged.

sdobson changed pull request status to merged

Sign up or log in to comment