running llama-stack container fails

[feloy]

When executing the command as described in the README:

podman run \
--net=host \
 --env RAMALAMA_URL=http://0.0.0.0:8080/ \         
 --env INFERENCE_MODEL=llama3.2:3b \
 quay.io/ramalama/llama-stack

I get the error:

[...]
INFO     2025-06-03 13:56:46,901 ramalama_stack.ramalama_adapter:66 inference: successfully connected to Ramalama at    
         `http://0.0.0.0:8080/`...                                                                                       
Traceback (most recent call last):
  File "/.venv/lib64/python3.13/site-packages/tiktoken/load.py", line 11, in read_file
    import blobfile
ModuleNotFoundError: No module named 'blobfile'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/server/server.py", line 597, in <module>
    main()
    ~~~~^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/server/server.py", line 487, in main
    impls = asyncio.run(construct_stack(config))
  File "/usr/lib64/python3.13/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ~~~~~~~~~~^^^^^^
  File "/usr/lib64/python3.13/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
  File "/usr/lib64/python3.13/asyncio/base_events.py", line 719, in run_until_complete
    return future.result()
           ~~~~~~~~~~~~~^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/stack.py", line 226, in construct_stack
    impls = await resolve_impls(run_config, provider_registry or get_provider_registry(run_config), dist_registry)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/resolver.py", line 143, in resolve_impls
    return await instantiate_providers(sorted_providers, router_apis, dist_registry, run_config)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/resolver.py", line 264, in instantiate_providers
    impl = await instantiate_provider(provider, deps, inner_impls, dist_registry, run_config)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/resolver.py", line 348, in instantiate_provider
    impl = await fn(*args)
           ^^^^^^^^^^^^^^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/routers/__init__.py", line 84, in get_auto_router_impl
    impl = api_to_routers[api.value](routing_table, **api_to_dep_impl)
  File "/.venv/lib64/python3.13/site-packages/llama_stack/distribution/routers/inference.py", line 78, in __init__
    self.tokenizer = Tokenizer.get_instance()
                     ~~~~~~~~~~~~~~~~~~~~~~^^
  File "/.venv/lib64/python3.13/site-packages/llama_stack/models/llama/llama3/tokenizer.py", line 51, in get_instance
    _INSTANCE = Tokenizer(os.path.join(os.path.dirname(__file__), "tokenizer.model"))
  File "/.venv/lib64/python3.13/site-packages/llama_stack/models/llama/llama3/tokenizer.py", line 63, in __init__
    mergeable_ranks = load_tiktoken_bpe(model_path)
  File "/.venv/lib64/python3.13/site-packages/tiktoken/load.py", line 148, in load_tiktoken_bpe
    contents = read_file_cached(tiktoken_bpe_file, expected_hash)
  File "/.venv/lib64/python3.13/site-packages/tiktoken/load.py", line 63, in read_file_cached
    contents = read_file(blobpath)
  File "/.venv/lib64/python3.13/site-packages/tiktoken/load.py", line 13, in read_file
    raise ImportError(
        "blobfile is not installed. Please install it by running `pip install blobfile`."
    ) from e
ImportError: blobfile is not installed. Please install it by running `pip install blobfile`.

The current latest tag is the same as the 0.9.0 one.

Note that it works correctly when I run it with a previous tag 0.8.5

[nathan-weinberg]

Seeing this in CI as well: https://github.com/containers/ramalama-stack/actions/runs/15415532994/job/43377322232

[nathan-weinberg]

Okay, this seems to be a result of this PR going into Llama Stack 0.2.8: llamastack/llama-stack#2271

Will fix this as part of #78