langflow-ai/langflow · error · HTTPException

No model provider is configured. Please configure at least o

Error message

No model provider is configured. Please configure at least one model provider in Settings.

What it means

The agentic assistant router resolves which model provider to use before executing a request. It builds the provider-variable mapping and queries enabled providers for the user; if the list is empty it raises HTTP 400 telling the user to configure at least one provider in Settings. Providers are enabled per-user via configured variables, so a fresh install or a user without API keys hits this immediately.

Source

Thrown at src/backend/base/langflow/agentic/api/router.py:70

    global_vars: dict[str, str]
    max_retries: int


async def _resolve_assistant_context(
    request: AssistantRequest,
    user_id: UUID,
    session: AsyncSession,
) -> _AssistantContext:
    """Resolve provider, model, API key, and build execution context.

    Raises:
        HTTPException: If provider is not configured or API key is missing.
    """
    provider_variable_map = get_model_provider_variable_mapping()
    enabled_providers, _ = await get_enabled_providers_for_user(user_id, session)

    if not enabled_providers:
        raise HTTPException(
            status_code=400,
            detail="No model provider is configured. Please configure at least one model provider in Settings.",
        )

    provider = request.provider
    if not provider:
        for preferred in PREFERRED_PROVIDERS:
            if preferred in enabled_providers:
                provider = preferred
                break
        if not provider:
            provider = enabled_providers[0]

    if provider not in enabled_providers:
        raise HTTPException(
            status_code=400,
            detail=f"Provider '{provider}' is not configured. Available providers: {enabled_providers}",
        )

View on GitHub (pinned to 976ec789d2)

Solutions

1. Open Settings and configure at least one model provider API key for the requesting user (creates the enabling variable).
2. As superuser, pre-seed provider variables for users via the Variables API/UI.
3. Verify with GET /agentic/check-config which providers are visible/enabled before sending assistant requests.

Example fix

# before
POST /api/v1/agentic/chat  # 400 No model provider is configured
# after: configure a key for the user, then
GET /api/v1/agentic/check-config  # shows provider enabled
POST /api/v1/agentic/chat

import requests
probe = requests.get("http://host/api/v1/agentic/check-config", headers=headers)
probe.raise_for_status()
providers = probe.json().get("providers") or probe.json().get("enabled_providers")
if not providers:
    show_settings_onboarding()  # don't fire the request that will 400

if resp.status_code == 400 and "No model provider" in resp.text:
    redirect_user_to_settings_page()

When it happens

Trigger: POST to any agentic assistant endpoint (chat/generate) for a user with no enabled providers — e.g. no OPENAI_API_KEY or ANTHROPIC_API_KEY variable configured for that user.

Common situations: First use of the Assistant after install; per-user global variables not yet set (keys set for a different user or only as server env without variable records); multi-user deployments where each user must configure keys.

Related errors

• Missing required configuration for {provider}: {', '.join(mi
• Provider '{provider}' is not configured. Available providers
• Unknown provider: {provider}
• This endpoint is not available
• Invalid path

AI-assisted analysis of langflow-ai/langflow@976ec789d2 (2026-08-14). Data as JSON: /api/errors/3fcffe1b8aada1f5. Report an issue: GitHub.