BerriAI/litellm · error · BedrockError

{err.response.text}

Error message

{err.response.text}

What it means

Synchronous Bedrock rerank POST returned non-2xx; the httpx.HTTPStatusError is re-raised as BedrockError with the upstream status code and raw response body ({err.response.text}). This is the sync twin of the async guard and surfaces the bedrock-agent-runtime error JSON verbatim.

Source

Thrown at litellm/llms/bedrock/rerank/handler.py:117

            return self.arerank(
                prepared_request,
                timeout=timeout,
                client=client if client is not None and isinstance(client, AsyncHTTPHandler) else None,
            )

        if client is None or not isinstance(client, HTTPHandler):
            client = _get_httpx_client()
        try:
            response: Final = client.post(
                url=prepared_request["endpoint_url"],
                headers=dict(prepared_request["prepped"].headers),
                data=prepared_request["body"],
                timeout=timeout,
            )
            response.raise_for_status()
        except httpx.HTTPStatusError as err:
            error_code: Final = err.response.status_code
            raise BedrockError(status_code=error_code, message=err.response.text)
        except httpx.TimeoutException:
            raise BedrockError(status_code=408, message="Timeout error occurred.")

        logging_obj.post_call(
            original_response=response.text,
            api_key="",
        )

        response_json: Final = response.json()

        return BedrockRerankConfig()._transform_response(response_json)

    def _prepare_request(
        self,
        model: str,
        api_base: str | None,
        extra_headers: dict | None,
        data: dict,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Parse the message body for the AWS error code and fix the named field/permission.
2. Confirm the rerank model ID and region; enable model access in the Bedrock console.
3. For 429, back off and retry; reduce requests per second.
4. Validate documents (non-empty strings/dicts) before the call.

try:
    result = litellm.rerank(model=model, query=q, documents=docs)
except BedrockError as e:
    if e.status_code == 400:
        raise InvalidRerankRequest(e.message) from e  # body has AWS ValidationException detail
    if e.status_code == 429:
        backoff_and_retry()
    raise

When it happens

Trigger: litellm.rerank(model='bedrock/<rerank-model>', ...) hitting ValidationException (bad params), ThrottlingException (429), AccessDeniedException (403), or ResourceNotFoundException (unknown model).

Common situations: Model access not enabled for the rerank model in the chosen region, oversized document lists, wrong model ID format, missing IAM permission on bedrock-agent-runtime.

Related errors

• err.response.text
• err.response.text
• APIError: {exception_provider} - {error_str}
• Timeout error occurred.
• Missing boto3 to call bedrock. Run 'pip install boto3'.

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/a67df0aafb57c05f. Report an issue: GitHub.