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.
Advisories
No advisories yet.
Fixes
Solution
No solution given by the vendor.
Workaround
No workaround given by the vendor.
References
History
Wed, 21 Oct 2026 00:45:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| Description | 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. | |
| Title | vLLM before 0.31.0 Stale allowed_token_ids Mask via InputBatch.condense | |
| First Time appeared |
Vllm
Vllm vllm |
|
| Weaknesses | CWE-440 | |
| CPEs | cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* | |
| Vendors & Products |
Vllm
Vllm vllm |
|
| References |
| |
| Metrics |
cvssV3_1
|
Projects
Sign in to view the affected projects.
Status: PUBLISHED
Assigner: VulnCheck
Published:
Updated: 2026-10-21T00:36:26.938Z
Reserved: 2026-10-20T21:41:06.180Z
Link: CVE-2026-114049
No data.
Status : Received
Published: 2026-10-21T01:17:27.513
Modified: 2026-10-21T01:17:27.513
Link: CVE-2026-114049
No data.
OpenCVE Enrichment
No data.
Weaknesses