4.2

CVE-2026-105755

vLLM: Flash late-interaction scoring caches query embeddings under a caller-controlled request id — cross-request integrity break and induced errors on `/score` and `/rerank`

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.
Daten sind bereitgestellt durch National Vulnerability Database (NVD)
Vllm ≫ Vllm Version < 0.30.0
VulnDex Vulnerability Enrichment
Diese Information steht angemeldeten Benutzern zur Verfügung. Login Login
Zu dieser CVE wurde keine Warnung gefunden.
EPSS Metriken
Typ Quelle Score Percentile
EPSS FIRST.org 0.2% 0.094
CVSS Metriken
Quelle Base Score Exploit Score Impact Score Vector String
security-advisories@github.com 4.2 1.6 2.5
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L
CWE-639 Authorization Bypass Through User-Controlled Key

The system's authorization functionality does not prevent one user from gaining access to another user's data or record by modifying the key value identifying the data.

https://github.com/vllm-project/vllm/releases/tag/v0.30.0
Release Notes
https://github.com/vllm-project/vllm/security/advisories/GHSA-2phq-3phc-84px
Vendor Advisory
https://github.com/vllm-project/vllm/pull/51445
Patch
Issue Tracking
https://github.com/vllm-project/vllm/commit/ee17d0d869203ef9a35ad73358a4987bba14b1fc
Patch