microsoft/autogen · error · ImportError

Dependencies for Llama Cpp not found. Please install llama-c

Error message

Dependencies for Llama Cpp not found. Please install llama-cpp-python: pip install autogen-ext[llama-cpp]

What it means

ImportError raised at package import time by autogen_ext.models.llama_cpp.__init__ when the underlying import of LlamaCppChatCompletionClient fails — i.e. the optional llama-cpp-python dependency is not installed. The exception chain (from e) preserves the original ImportError; the message tells you to install the extra via 'pip install autogen-ext[llama-cpp]'.

Source

Thrown at python/packages/autogen-ext/src/autogen_ext/models/llama_cpp/__init__.py:4

try:
    from ._llama_cpp_completion_client import LlamaCppChatCompletionClient
except ImportError as e:
    raise ImportError(
        "Dependencies for Llama Cpp not found. "
        "Please install llama-cpp-python: "
        "pip install autogen-ext[llama-cpp]"
    ) from e

__all__ = ["LlamaCppChatCompletionClient"]

View on GitHub (pinned to 027ecf0a37)

Solutions

1. pip install "autogen-ext[llama-cpp]" in the same environment/interpreter you run
2. If llama-cpp-python itself fails to build, install a prebuilt wheel matching your Python/ platform (e.g. pip install llama-cpp-python --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu) then reinstall the extra
3. Verify with: python -c "import llama_cpp; import autogen_ext.models.llama_cpp"

Example fix

# before
pip install autogen-ext
python -c "import autogen_ext.models.llama_cpp"  # ImportError

# after
pip install "autogen-ext[llama-cpp]"
python -c "import autogen_ext.models.llama_cpp"  # ok

try:
    import llama_cpp  # noqa: F401
except ImportError:
    raise SystemExit("Run: pip install 'autogen-ext[llama-cpp]'")

from autogen_ext.models.llama_cpp import LlamaCppChatCompletionClient

try:
    from autogen_ext.models.llama_cpp import LlamaCppChatCompletionClient
except ImportError as e:
    if "llama-cpp" in str(e):
        LlamaCppChatCompletionClient = None  # feature-flag local-llm support
    else:
        raise

When it happens

Trigger: import autogen_ext.models.llama_cpp (or any module importing it, e.g. LlamaCppChatCompletionClient references in configs) on an environment where the base autogen-ext package was installed without the [llama-cpp] extra; broken llama-cpp-python build (it compiles C++ and can fail to import even when pip-listed).

Common situations: requirements.txt lists only 'autogen-ext'; fresh venv/CI image without the extra; llama-cpp-python installed for a different Python or failed to compile against the local toolchain, leaving an importable-looking but broken install.

Related errors

• Dependencies for AzureAIAgent not found. Please install auto
• Dependencies for MagenticOneCoderAgent not found. Please ins
• Missing dependecies for DockerCommandLineCodeExecutor. Pleas
• Invalid name: {name}. Only letters, numbers, '_' and '-' are
• Invalid name: {name}. Name must be less than 64 characters.

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/327eceb4653c617b. Report an issue: GitHub.