Medium severity4.2NVD Advisory· Published Oct 5, 2026· Updated Nov 18, 2026
CVE-2026-105755
CVE-2026-105755
Description
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, flash late-interaction scoring at the /score and /rerank endpoints derives each worker's query_key value from the caller-controlled X-Request-Id header. A concurrent request that reuses a victim's identifier can overwrite the cached query embedding so the victim's documents are scored against the attacker's query, and shared use counters can also cause a late-interaction cache-miss error. This issue is fixed in version 0.30.0.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
vllmPyPI | < 0.30.0 | 0.30.0 |
Affected products
2Patches
Vulnerability mechanics
References
5- github.com/vllm-project/vllm/commit/ee17d0d869203ef9a35ad73358a4987bba14b1fcnvdPatchWEB
- github.com/vllm-project/vllm/pull/51445nvdIssue TrackingPatchWEB
- github.com/advisories/GHSA-2phq-3phc-84pxghsaADVISORY
- github.com/vllm-project/vllm/security/advisories/GHSA-2phq-3phc-84pxnvdVendor AdvisoryWEB
- github.com/vllm-project/vllm/releases/tag/v0.30.0nvdRelease NotesWEB
News mentions
1- VLLM: Batch of 11 Medium-Severity Vulnerabilities Disclosed TogetherVypr Intelligence · Oct 6, 2026