CVE-2026-105755

01/תיאור (מקור, אנגלית)

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.

02/מוצרים מושפעים

vllm: vllm

03/קישורים4

• github.com/vllm-project/vllm/security/advisories/GHSA-2phq-3ph‮… Vendor Advisory
• github.com/vllm-project/vllm/commit/ee17d0d869203ef9a35ad73358‮… Patch
• github.com/vllm-project/vllm/pull/51445 Issue TrackingPatch
• github.com/vllm-project/vllm/releases/tag/v0.30.0 Release Notes