CVE-2026-114049
CVE-2026-114049 · vLLM before 0.31.0 Stale allowed_token_ids Mask via InputBatch.condense
This page contains facts from public vulnerability sources. Check the original advisory before updating.
vLLM before 0.31.0 contains a cross-request mask leak in InputBatch.condense() that allows remote attackers to constrain other clients' outputs by leaving stale allowed_token_ids rows. Attackers can submit concurrent constrained generate requests, including via unauthenticated /invocations, so recycled batch slots apply their token allowlist to victims' unrestricted requests.
CVSS 6.3 · Medium · EPSS 0.3% · Fix available
- Published
- Updated
Affected products and versions
vllm
- Versions
- ≥ 0, < 0.31.0: Affected (semver)
- 0.31.0: Not affected (semver)
Update to 0.31.0