BerriAI/litellm · error · Exception

Invalid api base. api_base=None. Set in call or via `INFINIT

Error message

Invalid api base. api_base=None. Set in call or via `INFINITY_API_BASE` env var.

What it means

API-base resolution guard for the Infinity rerank provider: no api_base was supplied and INFINITY_API_BASE is unset, so the self-hosted Infinity rerank endpoint location is unknown.

Source

Thrown at litellm/rerank_api/main.py:250

                timeout=optional_params.timeout,
                api_key=dynamic_api_key or optional_params.api_key,
                api_base=api_base,
                _is_async=_is_async,
                headers=headers or litellm.headers or {},
                client=client,
                model_response=model_response,
                litellm_params=rerank_litellm_params,
            )
        elif _custom_llm_provider == litellm.LlmProviders.INFINITY:
            # Implement Infinity rerank logic
            api_key = dynamic_api_key or optional_params.api_key or litellm.api_key

            api_base = (
                dynamic_api_base or optional_params.api_base or litellm.api_base or get_secret_str("INFINITY_API_BASE")
            )

            if api_base is None:
                raise Exception("Invalid api base. api_base=None. Set in call or via `INFINITY_API_BASE` env var.")

            response = base_llm_http_handler.rerank(
                model=model,
                custom_llm_provider=_custom_llm_provider,
                provider_config=rerank_provider_config,
                optional_rerank_params=optional_rerank_params,
                logging_obj=litellm_logging_obj,
                timeout=optional_params.timeout,
                api_key=dynamic_api_key or optional_params.api_key,
                api_base=api_base,
                _is_async=_is_async,
                headers=headers or litellm.headers or {},
                client=client,
                model_response=model_response,
                litellm_params=rerank_litellm_params,
            )
        elif _custom_llm_provider == litellm.LlmProviders.TOGETHER_AI:
            # Implement Together AI rerank logic

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Pass api_base in the call or set the INFINITY_API_BASE environment variable.

When it happens

Trigger: Thrown at litellm/rerank_api/main.py:250 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/c6a941e4dae4d881. Report an issue: GitHub.