BerriAI/litellm · error · HTTPException

Team rpm_limit ({data.rpm_limit}) exceeds organization's rpm

Error message

Team rpm_limit ({data.rpm_limit}) exceeds organization's rpm_limit ({org_table.litellm_budget_table.rpm_limit}). Organization: {org_table.organization_id}

What it means

Hierarchical budget validation in the new-team endpoint (/team/new). The requested team's rpm_limit is larger than the rpm_limit set on the parent organization's litellm_budget_table, which would violate LiteLLM's org-to-team limit containment. Fires only when the parent org has an rpm_limit configured and the new team exceeds it. Lower the team rpm_limit or raise the organization's limit first.

Source

Thrown at litellm/proxy/management_endpoints/team_endpoints.py:986

        data.tpm_limit is not None
        and org_table.litellm_budget_table is not None
        and org_table.litellm_budget_table.tpm_limit is not None
        and data.tpm_limit > org_table.litellm_budget_table.tpm_limit
    ):
        raise HTTPException(
            status_code=400,
            detail={
                "error": f"Team tpm_limit ({data.tpm_limit}) exceeds organization's tpm_limit ({org_table.litellm_budget_table.tpm_limit}). Organization: {org_table.organization_id}"
            },
        )

    if (
        data.rpm_limit is not None
        and org_table.litellm_budget_table is not None
        and org_table.litellm_budget_table.rpm_limit is not None
        and data.rpm_limit > org_table.litellm_budget_table.rpm_limit
    ):
        raise HTTPException(
            status_code=400,
            detail={
                "error": f"Team rpm_limit ({data.rpm_limit}) exceeds organization's rpm_limit ({org_table.litellm_budget_table.rpm_limit}). Organization: {org_table.organization_id}"
            },
        )

    # Check guaranteed throughput limits (only if applicable)
    rpm_limit_type: Final = getattr(data, "rpm_limit_type", None) or (
        data.metadata.get("rpm_limit_type", None) if data.metadata else None
    )
    tpm_limit_type: Final = getattr(data, "tpm_limit_type", None) or (
        data.metadata.get("tpm_limit_type", None) if data.metadata else None
    )

    if tpm_limit_type != "guaranteed_throughput" and rpm_limit_type != "guaranteed_throughput":
        return
    # get all organization teams
    # calculate allocated tpm/rpm limit

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Set the team rpm_limit at or below the organization's rpm_limit.

When it happens

Trigger: Thrown at litellm/proxy/management_endpoints/team_endpoints.py:986 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/01e2d5381d8334d5. Report an issue: GitHub.