bytedance/deer-flow · critical · ValueError

No chat models are configured. Please configure at least…

Error message

No chat models are configured. Please configure at least one model in config.yaml.

What it means

ValueError from _resolve_model_name in the lead agent: app_config.models is empty, so there is no default model to fall back to. The agent refuses to build rather than running modelless. This is a configuration error — config.yaml has zero entries under models.

Solutions

Add at least one full model entry under models: in config.yaml and restart the Gateway.
Validate config with `make doctor` (or the config validation endpoint) to catch YAML structure errors.
If config is managed via API, re-fetch it and confirm models is a non-empty list.
For tests, inject an AppConfig stub with at least one model instead of relying on global config.

Example fix

# config.yaml
# before
models: []

# after
models:
  - name: gpt-4o
    provider: openai
    api_key: ${OPENAI_API_KEY}

Defensive patterns

Strategy: validation

Validate before calling

from deerflow.config import get_app_config
cfg = get_app_config()
assert cfg.models, 'config.yaml has no models configured — agent creation will fail'

Type guard

def has_model(cfg) -> bool:
    return bool(getattr(cfg, 'models', None)) and all(getattr(m, 'name', None) for m in cfg.models)

Try / catch

try:
    agent = create_agent(...)
except ValueError as e:
    if 'No chat models are configured' in str(e):
        raise SystemExit('Fix config.yaml: add at least one model entry') from e
    raise

Prevention

Run `make doctor` after config changes.
Fail startup early: validate models non-empty at Gateway boot, not at first run.
Add a config schema test in CI that loads the example config and asserts models is non-empty.

When it happens

Trigger: Any agent/run creation when config.yaml (or the resolved AppConfig) contains no models: empty models list, failed config load falling through to defaults, or a config override wiping the list.

Common situations: Fresh clone where config.yaml was copied from example but models section left empty; CI/test environment without model config; typo in YAML indentation making models parse as a scalar; config API PATCH that replaced models with [].

Related errors

AI-assisted analysis of bytedance/deer-flow@1dd6ba1acb (2026-08-14). Data as JSON: /api/errors/14123b5f4a6cda87. Report an issue: GitHub.

Appendix: source

Thrown at backend/packages/harness/deerflow/agents/lead_agent/agent.py:133

        tools.append(memory_tool)
        existing_names.add(memory_tool.name)


def _get_runtime_config(config: RunnableConfig) -> dict:
    """Merge legacy configurable options with LangGraph runtime context."""
    cfg = dict(config.get("configurable", {}) or {})
    context = config.get("context", {}) or {}
    if isinstance(context, dict):
        cfg.update(context)
    return cfg


def _resolve_model_name(requested_model_name: str | None = None, *, app_config: AppConfig | None = None) -> str:
    """Resolve a runtime model name safely, falling back to default if invalid. Returns None if no models are configured."""
    app_config = app_config or get_app_config()
    default_model_name = app_config.models[0].name if app_config.models else None
    if default_model_name is None:
        raise ValueError("No chat models are configured. Please configure at least one model in config.yaml.")

    if requested_model_name and app_config.get_model_config(requested_model_name):
        return requested_model_name

    if requested_model_name and requested_model_name != default_model_name:
        logger.warning(f"Model '{requested_model_name}' not found in config; fallback to default model '{default_model_name}'.")
    return default_model_name


def _authorize_model_name(
    model_name: str,
    *,
    context: Mapping[str, Any],
    app_config: AppConfig,
) -> str:
    """Enforce ``model:use`` authorization on the resolved model name.

    When ``authorization.enabled`` is false this is a no-op (returns

View on GitHub (pinned to 1dd6ba1acb)