odysseus-dev/odysseus · error · HTTPException

Could not auto-detect provider. Pass base_url (e.g. 'https:/

Error message

Could not auto-detect provider. Pass base_url (e.g. 'https://api.deepseek.com/v1') or provider ('deepseek', 'openai', 'groq', etc.)

What it means

HTTP 400 from POST /v1/chat: no base_url was supplied and _resolve_base_url(model, provider) could not map the request to a known provider. Resolution keys off the model name prefix or an explicit provider string against KNOWN_PROVIDERS (deepseek, openai, groq, ...).

Source

Thrown at routes/webhook/webhook_routes.py:296

        # --- Case 2: Direct API key + model (no pre-configured endpoint needed) ---
        if not sess and body.api_key:
            api_key = body.api_key.strip()
            model = body.model or "deepseek-chat"

            # Validate only token-supplied direct base_url; auto-resolved known-provider
            # URLs are not subject to extra local/LAN blocking beyond existing provider logic.
            direct_base_url = body.base_url.strip().rstrip("/") if body.base_url else None
            if direct_base_url:
                try:
                    base_url = validate_public_http_url(direct_base_url)
                except ValueError as e:
                    detail = str(e).replace("URL", "base_url", 1)
                    raise HTTPException(400, detail)
            else:
                base_url = _resolve_base_url(model, body.provider)
            if not base_url:
                raise HTTPException(400,
                    "Could not auto-detect provider. Pass base_url (e.g. 'https://api.deepseek.com/v1') "
                    "or provider ('deepseek', 'openai', 'groq', etc.)")
            base_url = normalize_base(base_url)
            endpoint_url = build_chat_url(base_url)

            if not session_manager:
                raise HTTPException(500, "Session manager not available")

            sid = str(uuid.uuid4())
            sess = session_manager.create_session(
                session_id=sid, name="API Chat", endpoint_url=endpoint_url,
                model=model, owner=token_owner,
            )
            sess.headers = build_headers(api_key, base_url)
            session_manager.save_sessions()
            session_id = sid

        # --- Case 3: Fall back to first configured ModelEndpoint ---

View on GitHub (pinned to f9235ebbf1)

Solutions

1. Add an explicit provider field matching a KNOWN_PROVIDERS key (e.g. 'deepseek', 'openai', 'groq')
2. Or supply a public base_url explicitly (see the SSRF constraints of error 929)
3. Or configure a ModelEndpoint in Admin and call with no api_key (Case 3 fallback)

Example fix

// before
{"message":"hi","api_key":"sk-...","model":"llama-3-70b"}

// after
{"message":"hi","api_key":"sk-...","model":"llama-3-70b","provider":"groq"}

KNOWN = {"deepseek", "openai", "groq"}  # keep in sync with KNOWN_PROVIDERS
if not base_url and (provider not in KNOWN and not model.startswith(tuple(KNOWN))):
    require_explicit_provider_or_base_url()

if resp.status_code == 400 and 'auto-detect provider' in detail:
    retry_with(provider=guess_provider(model), base_url=None)

When it happens

Trigger: Passing api_key + an unrecognized model (e.g. 'llama-3-70b') with no base_url and no provider; passing a provider string not in KNOWN_PROVIDERS (typo like 'opena1'); provider given as None and model name doesn't start with any known prefix.

Common situations: Using a third-party OpenAI-compatible aggregator (OpenRouter, Together) whose model names don't match known prefixes; new model names the resolver's prefix table doesn't know yet.

Related errors

• No session, api_key, or configured endpoints. Pass api_key +
• Message is required
• detail
• await res.text()
• HTTP ${res.status}

AI-assisted analysis of odysseus-dev/odysseus@f9235ebbf1 (2026-08-14). Data as JSON: /api/errors/459945794bb99e59. Report an issue: GitHub.