Disclosed weakness · no CVE recorded

LiteLLM queued chat requests could skip configured budget and request checks

LiteLLM · LiteLLM Proxy · queued chat completions

First documented disclosure · Review checkpoint

Summary

Queued chat calls could continue after spend limits were reached or omit information that normal chat requests had to provide.

Severity and exploitation evidence

No CVSS score recorded in this review.

Exploitation not established: The maintainer reports controlled checks comparing the two routes. The cited sources do not report malicious use.

Exploitation evidence source

Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.

How the weakness works

The queued route omitted admission checks, budget reservations and required tag handling applied to ordinary chat completions.

Conditions required

A proxy exposes queued chat to a caller and relies on budget, identity-field or required-tag rules to restrict requests.

Potential impact

A caller could consume model capacity beyond assigned budgets. Usage records could also omit tags used to attribute spending.

Affected versions and configurations

The public fix identifies the affected implementation. The full affected version range is unspecified.

Fix and mitigation status

LiteLLM lists the fix in its development release 1.111.0-dev.4.

Practical next steps

Deploy the queued-route fix or limit access to queued chat until its controls are verified. Review queued usage against configured budgets.

Operating lesson

Apply the same admission rules to each path that can invoke a billed model.

Evidence and limits

The maintainer documents a correction. The cited sources provide no issue-specific severity rating or CVE.

Sources