| ID |
CVE-2026-105755
|
| Sažetak |
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. |
| Reference |
|
| CVSS |
| Base: | 4.2 |
| Impact: | 2.5 |
| Exploitability: | 1.6 |
|
| Pristup |
| Vektor | Složenost | Autentikacija |
| NETWORK |
HIGH |
LOW |
|
| Impact |
| Povjerljivost | Cjelovitost | Dostupnost |
| NONE |
LOW |
LOW |
|
| CVSS vektor |
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L |
| Zadnje važnije ažuriranje |
05-10-2026 - 23:17 |
| Objavljeno |
05-10-2026 - 23:17 |
