BerriAI/litellm · error · Timeout

BedrockException: Timeout Error

Error message

BedrockException: Timeout Error - {error_str}

What it means

LiteLLM throws Timeout when Bedrock connectivity or inference exceeds time limits: 'Connect timeout on endpoint URL' (botocore failing to establish the TCP/TLS connection) or 'timed out' (request exceeded the client's timeout). The mapper normalizes both into litellm.Timeout so callers get one uniform type with the model attached. Unlike string-mapped branches, no response object is attached.

Solutions

1. Raise the timeout: litellm.completion(..., timeout=300) or request_timeout, and set num_retries to ride out transient slowness
2. Switch to streaming for long outputs so time-to-first-token governs instead of full-generation wall clock
3. Verify network egress to bedrock-runtime.<region>.amazonaws.com:443 and corporate proxy configuration
4. Check DNS / VPC endpoint resolution if connect timeouts (not read timeouts) recur

Example fix

# before
resp = litellm.completion(model="bedrock/...", messages=msgs)  # Connect timeout on endpoint URL

# after
resp = litellm.completion(model="bedrock/...", messages=msgs, timeout=300, num_retries=2, stream=True)

import litellm

def is_timeout(e: BaseException) -> bool:
    return isinstance(e, litellm.Timeout)

try:
    resp = litellm.completion(model='bedrock/...', messages=msgs, timeout=300)
except litellm.Timeout:
    resp = litellm.completion(model='bedrock/...', messages=msgs, timeout=300, stream=True, num_retries=2)

When it happens

Trigger: Slow non-streaming generations exceeding the configured litellm timeout (default 600s); network egress blocked to bedrock-runtime.<region>.amazonaws.com:443; DNS/proxy/VPC-endpoint misconfiguration producing botocore ConnectTimeoutError.

Common situations: Corporate proxies that must allow-list AWS endpoints; long max_tokens generations without streaming; VPC endpoints with broken private DNS; aggressive per-request timeouts set via timeout=/request_timeout.

Understand the failure class

Background: Request timed out: what client-side request timeouts mean across libraries (Request timed out, TIMED_OUT, APITimeoutError) — this error's family across 39 libraries.

• Timeouts: ETIMEDOUT, deadlines, and hung requests — what actually expires when a request times out.

Related errors

• Timeout error occurred.
• Timeout error occurred.
• Timeout error occurred.
• Cisco AI Defense API call timed out after s
• Error fetching prompt version

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/9bfbff13dcbd0dd3. Report an issue: GitHub.

Appendix: source

Thrown at litellm/litellm_core_utils/exception_mapping_utils.py:883

            llm_provider="bedrock",
            response=getattr(original_exception, "response", None),
        )
    elif "AccessDeniedException" in error_str:
        raise PermissionDeniedError(
            message=f"BedrockException PermissionDeniedError - {error_str}",
            model=model,
            llm_provider="bedrock",
            response=getattr(original_exception, "response", None),
        )
    elif "throttlingException" in error_str or "ThrottlingException" in error_str:
        raise RateLimitError(
            message=f"BedrockException: Rate Limit Error - {error_str}",
            model=model,
            llm_provider="bedrock",
            response=getattr(original_exception, "response", None),
        )
    elif "Connect timeout on endpoint URL" in error_str or "timed out" in error_str:
        raise Timeout(
            message=f"BedrockException: Timeout Error - {error_str}",
            model=model,
            llm_provider="bedrock",
        )
    elif "Could not process image" in error_str:
        raise litellm.InternalServerError(
            message=f"BedrockException - {error_str}",
            model=model,
            llm_provider="bedrock",
        )
    elif hasattr(original_exception, "status_code"):
        if original_exception.status_code == 500:
            raise ServiceUnavailableError(
                message=f"BedrockException - {original_exception.message}",
                llm_provider="bedrock",
                model=model,
                response=httpx.Response(
                    status_code=500,

View on GitHub (pinned to 77b7c6c40c)