langchain-ai/deepagents · error · ModelConfigError

Could not import module '{module_path}' for provider '{provi

Error message

Could not import module '{module_path}' for provider '{provider}': {e}

What it means

Raised when the module portion of a configured `class_path` cannot be imported; the ImportError is wrapped as `ModelConfigError` with the original exception chained via `from e`, preserving the root cause. A module that itself raises ImportError during import also lands here.

Source

Thrown at libs/code/deepagents_code/config.py:5386

        BaseChatModel as _BaseChatModel,  # Runtime import; module level is typing only
    )

    from deepagents_code.model_config import ModelConfigError

    if ":" not in class_path:
        msg = (
            f"Invalid class_path '{class_path}' for provider '{provider}': "
            "must be in module.path:ClassName format"
        )
        raise ModelConfigError(msg)

    module_path, class_name = class_path.rsplit(":", 1)

    try:
        module = importlib.import_module(module_path)
    except ImportError as e:
        msg = f"Could not import module '{module_path}' for provider '{provider}': {e}"
        raise ModelConfigError(msg) from e

    cls = getattr(module, class_name, None)
    if cls is None:
        msg = (
            f"Class '{class_name}' not found in module '{module_path}' "
            f"for provider '{provider}'"
        )
        raise ModelConfigError(msg)

    if not (isinstance(cls, type) and issubclass(cls, _BaseChatModel)):
        msg = (
            f"'{class_path}' is not a BaseChatModel subclass (got {type(cls).__name__})"
        )
        raise ModelConfigError(msg)

    try:
        return cls(model=model_name, **kwargs)
    except Exception as e:

View on GitHub (pinned to a1af029e6e)

Solutions

1. Read the chained ImportError in the message — it names the real cause
2. Install the module's package into the same environment dcode runs in (`pip install my_pkg` or `uv pip install -e ./my_pkg`)
3. Fix the module path in config.toml or the missing transitive dependency

Example fix

// before (config.toml)
class_path = "my_pkd.models:MyChatModel"  # typo

// after
class_path = "my_pkg.models:MyChatModel"
$ pip install -e ./my_pkg

import importlib.util
def module_importable(module_path: str) -> bool:
    try:
        return importlib.util.find_spec(module_path) is not None
    except (ImportError, ValueError):
        return False

from deepagents_code.model_config import ModelConfigError
try:
    model = create_model(spec, class_path=class_path)
except ModelConfigError as e:
    if "Could not import module" in str(e):
        print(f"Install the package defining {class_path.split(':')[0]}")
        raise SystemExit(1)
    raise

When it happens

Trigger: `_create_from_class` calls `importlib.import_module(module_path)` which raises ImportError — the package is not installed in the active environment, the module name is misspelled, or a transitive import inside the module fails (config.py:5382-5386).

Common situations: Custom model package installed in a different virtualenv than dcode's; typo in the module path (`my_pkd.models`); a dependency of the module missing after an upgrade; editable install stale after a rename.

Related errors

• Missing dependencies for '{provider}' sandbox. {install_hint
• Class '{class_name}' not found in module '{module_path}'
• Package {package_root} is not installed
• modes can only be provided when agent is a factory
• models can only be provided when agent is a factory

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/11da187a7f05d814. Report an issue: GitHub.