BerriAI/litellm · error · HTTPException

Allocated RPM limit={model_specific_rpm_limit.get(model, 0)}

Error message

Allocated RPM limit={model_specific_rpm_limit.get(model, 0)} + Team RPM limit={rpm_limit} is greater than {entity_type} RPM limit={entity_rpm_limit}

What it means

Model-level RPM over-allocation check when creating/updating a team under a parent entity: summing the per-model rpm limits already allocated to sibling teams plus the new team's request exceeds the entity's RPM ceiling, so a 400 blocks the allocation. At-fault input is model_rpm_limit pushing the total over the parent's limit.

Source

Thrown at litellm/proxy/management_endpoints/team_endpoints.py:794

    if model_rpm_limit is None and model_tpm_limit is None:
        return

    # get total model specific tpm/rpm limit
    model_specific_rpm_limit: Final[dict[str, int]] = {}
    model_specific_tpm_limit: Final[dict[str, int]] = {}

    for team in teams:
        if team.metadata and team.metadata.get("model_rpm_limit", None) is not None:
            for model, rpm_limit in team.metadata.get("model_rpm_limit", {}).items():
                model_specific_rpm_limit[model] = model_specific_rpm_limit.get(model, 0) + rpm_limit
        if team.metadata and team.metadata.get("model_tpm_limit", None) is not None:
            for model, tpm_limit in team.metadata.get("model_tpm_limit", {}).items():
                model_specific_tpm_limit[model] = model_specific_tpm_limit.get(model, 0) + tpm_limit

    if model_rpm_limit is not None:
        for model, rpm_limit in model_rpm_limit.items():
            if entity_rpm_limit is not None and model_specific_rpm_limit.get(model, 0) + rpm_limit > entity_rpm_limit:
                raise HTTPException(
                    status_code=400,
                    detail=f"Allocated RPM limit={model_specific_rpm_limit.get(model, 0)} + Team RPM limit={rpm_limit} is greater than {entity_type} RPM limit={entity_rpm_limit}",
                )
            elif entity_model_rpm_limit_dict:
                entity_model_specific_rpm_limit = entity_model_rpm_limit_dict.get(model)
                if (
                    entity_model_specific_rpm_limit
                    and model_specific_rpm_limit.get(model, 0) + rpm_limit > entity_model_specific_rpm_limit
                ):
                    raise HTTPException(
                        status_code=400,
                        detail=f"Allocated RPM limit={model_specific_rpm_limit.get(model, 0)} + Team RPM limit={rpm_limit} is greater than {entity_type} RPM limit={entity_model_specific_rpm_limit}",
                    )

    if model_tpm_limit is not None:
        for model, tpm_limit in model_tpm_limit.items():
            if entity_tpm_limit is not None and model_specific_tpm_limit.get(model, 0) + tpm_limit > entity_tpm_limit:
                raise HTTPException(

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Lower the model-specific or team RPM limit so the sum does not exceed the entity RPM limit.

When it happens

Trigger: Thrown at litellm/proxy/management_endpoints/team_endpoints.py:794 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/9597ecf3d87457e1. Report an issue: GitHub.