CVE-2026-105755
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.
Does this matter?
Lower severity and a low EPSS score (0.20%). Track it; it rarely justifies an emergency change on its own.
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.
- CVSS 3.1
- 4.2 MEDIUMCVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L
- EPSS
- 0.20% probability · 10th percentile
- Public exploits
- None in Exploit-DB
- CISA KEV
- Not listed
- Weakness
- CWE-639
- Affected
- vllm/vllm
- Source
- security-advisories@github.com
EPSS moved from 0.2% on 6 October 2026 to 0.2% on 26 October 2026.
References
- https://github.com/vllm-project/vllm/commit/ee17d0d869203ef9a35ad73358a4987bba14b1fcPatch
- https://github.com/vllm-project/vllm/pull/51445Issue Tracking, Patch
- https://github.com/vllm-project/vllm/releases/tag/v0.30.0Release Notes
- https://github.com/vllm-project/vllm/security/advisories/GHSA-2phq-3phc-84pxVendor Advisory
Source: NVD record, EPSS from FIRST.org, KEV from CISA, exploits from Exploit-DB. Refreshed daily. Download this record as JSON.