Deployment-level TPM enforcement per-pod breaks effective rate limits

What this error means

tpm_limit × N_replica — deployment-level tokens-per-minute enforcement is per-pod, not cross-pod consistent is a LiteLLM failure pattern reported for developers trying to understand why litellm tpm rate limits don’t aggregate across pods and how to enforce correct per-deployment token limits. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.

Why this happens

GitHub issue BerriAI/litellm#27736 by hula-la (May 12 2026). LiteLLM proxy sets tpm_limit at deployment level but enforcement happens per-replica pod, meaning a 4-replica deployment effectively has 4× the intended rate limit. This bypasses cost controls and can trigger upstream provider rate limits unexpectedly. Labels: bug, proxy. Critical for teams using LiteLLM proxy for budget management across distributed deployments.

Common causes

• GitHub issue BerriAI/litellm#27736 by hula-la (May 12 2026). LiteLLM proxy sets tpm_limit at deployment level but enforcement happens per-replica pod, meaning a 4-replica deployment effectively has 4× the intended rate limit. This bypasses cost controls and can trigger upstream provider rate limits unexpectedly. Labels: bug, proxy. Critical for teams using LiteLLM proxy for budget management across distributed deployments.

Quick fixes

1. Confirm the exact error signature matches tpm_limit × N_replica — deployment-level tokens-per-minute enforcement is per-pod, not cross-pod consistent.
2. Check the LiteLLM account, local tool state, and provider configuration involved in the failing workflow.
3. Reduce request pressure, check quota or plan limits, and retry with backoff instead of immediate repeated requests.

Platform/tool-specific checks

• Verify the command, editor, extension, or API client that produced the error.
• Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
• Avoid deleting credentials, local model data, or project settings until the failing scope is clear.

Step-by-step troubleshooting

1. Capture the exact error message and the command, editor action, or request that triggered it.
2. Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
3. Review the source evidence below and compare it with your environment.
4. Apply one change at a time and rerun the smallest failing action.
5. Keep the working fix documented for the team or deployment environment.

How to prevent it

• Keep provider/tool configuration documented.
• Record non-secret diagnostics such as tool version, provider name, model name, and command path.
• Add a lightweight check before CI or production workflows depend on the tool.

Sources checked

• https://github.com/BerriAI/litellm/issues/27736

Evidence note: GitHub issue BerriAI/litellm#27736 by hula-la (May 12 2026). LiteLLM proxy sets tpm_limit at deployment level but enforcement happens per-replica pod, meaning a 4-replica deployment effectively has 4× the intended rate limit. This bypasses cost controls and can trigger upstream provider rate limits unexpectedly. Labels: bug, proxy. Critical for teams using LiteLLM proxy for budget management across distributed deployments.

Related errors

• LiteLLM

FAQ

What should I check first?

Start with the exact tpm_limit × N_replica — deployment-level tokens-per-minute enforcement is per-pod, not cross-pod consistent text and the smallest action that reproduces it.

Can I ignore this error?

No. Treat it as a failed LiteLLM workflow until the root cause is understood.

Is this guaranteed to have one fix?

No. The imported evidence supports the troubleshooting path above, but tool behavior can vary by account, plan, version, provider, and local configuration.

How do I know the fix worked?

Rerun the same command, editor action, or request. The fix is working when that action completes without tpm_limit × N_replica — deployment-level tokens-per-minute enforcement is per-pod, not cross-pod consistent.