CVE-2026-114049 — vLLM before 0.31.0 Stale allowed_token_ids Mask via InputBatch.condense
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_id…
Published: 2026-10-21 · CVSS: 6.3 · CVSS 4.0
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.
Source: CVE Program · Revision 8b16575c39f1e1810c5e863010da073d2d3285e7