langchain-ai/deepagents · error · UnknownProviderError

Unable to infer a model provider for {model_spec!r}. Specify

Error message

Unable to infer a model provider for {model_spec!r}. Specify one explicitly (e.g. 'anthropic:{model_spec}') or see the provider reference at {docs_url}.

What it means

Raised when `create_model` cannot determine which provider to use for a model spec. Both the app's auto-detection (`detect_provider`) and `init_chat_model`'s own inference failed, so a structured `UnknownProviderError` is raised so the UI can render the provider-reference docs URL as a clickable link.

Source

Thrown at libs/code/deepagents_code/config.py:5506

            else:
                if hint.command is not None:
                    install_hint = f"Install with: {hint.command}"
                else:
                    install_hint = f"Install the '{package}' package manually"
                msg = (
                    f"Missing package for provider '{provider}'. "
                    f"{install_hint}, then retry with `/model`."
                )
            raise MissingProviderPackageError(
                msg, provider=provider, package=package
            ) from e
        raise ModelConfigError(msg) from e
    except (ValueError, TypeError) as e:
        if not provider:
            # Both app auto-detection and `init_chat_model`'s own inference
            # failed; surface a structured error so the UI can render the
            # docs URL as a clickable link.
            raise UnknownProviderError(model_spec=model_name) from e
        spec = f"{provider}:{model_name}"
        msg = f"Invalid model configuration for '{spec}': {e}"
        raise ModelConfigError(msg) from e
    except Exception as e:  # provider SDK auth/network errors
        spec = f"{provider}:{model_name}" if provider else model_name
        msg = f"Failed to initialize model '{spec}': {e}"
        raise ModelConfigError(msg) from e


@dataclass(frozen=True)
class ModelResult:
    """Result of creating a chat model, bundling the model with its metadata.

    This separates model creation from runtime-state mutation so callers can
    decide when to commit the metadata to process-wide state.

    Attributes:
        model: The instantiated chat model.

View on GitHub (pinned to a1af029e6e)

Solutions

Add an explicit provider prefix: use 'anthropic:my-model' instead of 'my-model'.
Set a default model spec in config.toml or the relevant env var so `create_model()` has a fully qualified default.
Check the model name for typos against the provider's official model list.
Install the provider package (e.g. langchain-anthropic) — missing packages can prevent inference.

Example fix

// before
model = create_model("my-weird-model-id")
// after
model = create_model("anthropic:my-weird-model-id")

Defensive patterns

Strategy: validation

Validate before calling

def has_provider_hint(spec: str) -> bool:
    if ":" in spec:
        return True
    known_prefixes = ("gpt-", "o1", "o3", "claude", "gemini", "grok", "command", "bedrock/")
    return spec.lower().startswith(known_prefixes)
# call create_model only when has_provider_hint(spec), else pass "provider:" + spec

Try / catch

try:
    result = create_model(spec)
except UnknownProviderError:
    result = create_model(f"anthropic:{spec}")  # or prompt user for provider

Prevention

Always qualify non-standard model ids with an explicit 'provider:' prefix
Set a default model spec in config.toml so bare create_model() never needs inference
Keep the provider package installed so init_chat_model inference works
Verify model names against the provider's official docs

When it happens

Trigger: Calling `create_model()` or `create_model(model_spec)` with a bare model name (no `provider:` prefix) whose name matches no known prefix, no default model is configured, and no provider can be inferred from environment credentials — and `init_chat_model` also raises ValueError/TypeError during inference.

Common situations: Typos in model names ('claud-sonnet-4-5'), obscure or custom model IDs not in `detect_provider`'s prefix list, missing default in config.toml or env, using a new provider model before the app knows its prefix.

Related errors

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/dc7f9908e73060d1. Report an issue: GitHub.