Expected Behavior Violation Affecting vllm package, versions [,0.31.0)
Threat Intelligence
The probability is the direct output of the EPSS model, and conveys an overall sense of the threat of exploitation in the wild. The percentile measures the EPSS probability relative to all known EPSS scores. Note: This data is updated daily, relying on the latest available EPSS model version. Check out the EPSS documentation for more details.
Do your applications use this vulnerable package?
In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.
Test your applications- Snyk IDSNYK-PYTHON-VLLM-21347782
- published21 Oct 2026
- disclosed21 Oct 2026
- creditSunny Rangnani
Introduced: 21 Oct 2026
NewCVE-2026-114049 (opens in a new tab)How to fix?
Upgrade vllm to version 0.31.0 or higher.
Overview
vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs
Affected versions of this package are vulnerable to Expected Behavior Violation in the condense method of vllm/v1/worker/gpu_input_batch.py, where stale allowed_token_ids mask entries are not cleared after a request slot is vacated. When the InputBatch is condensed to remove completed requests, the mask tensor entry at the vacated index retains the token-id allowlist from the previous occupant, causing a subsequent request assigned to that slot to inherit incorrect token constraints and produce outputs that violate its intended generation policy.