Advisory details:
pypa / vllm/PYSEC-2026-4198
| Advisory ID | pypa/vllm/PYSEC-2026-4198 |
| Aliases |
CVE-2026-105755
GHSA-2phq-3phc-84px PYSEC-2026-4198 |
| Summary | 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. |
| Status | Published |
| Exploitability | 0.5 |
| Weighted Severity | 3.8 |
| Risk | 1.9 |
| Affected and Fixed Packages | Package Details |
| Introduced and Fixed Package Commit Patches | Package Commit Patches Details |
| There are no known CWE. |
| System | Score | Found at |
|---|---|---|
| cvssv3.1 | 4.2 | https://github.com/pypa/advisory-database/blob/main/vulns/vllm/PYSEC-2026-4198.yaml |
No EPSS data available for this advisory.
| Patch URL | |
|---|---|
| There are no known patches. | |
| Attack Vector (AV) | Attack Complexity (AC) | Privileges Required (PR) | User Interaction (UI) | Scope (S) | Confidentiality Impact (C) | Integrity Impact (I) | Availability Impact (A) |
|---|---|---|---|---|---|---|---|
network adjacent_network local physical |
low high |
none low high |
none required |
unchanged changed |
high low none |
high low none |
high low none |
There are no SSVC decisions available.