CVE-2026-22778

{"id": "CVE-2026-22778", "cveTags": [], "metrics": {"cvssMetricV31": [{"type": "Secondary", "source": "security-advisories@github.com", "cvssData": {"scope": "UNCHANGED", "version": "3.1", "baseScore": 9.8, "attackVector": "NETWORK", "baseSeverity": "CRITICAL", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H", "integrityImpact": "HIGH", "userInteraction": "NONE", "attackComplexity": "LOW", "availabilityImpact": "HIGH", "privilegesRequired": "NONE", "confidentialityImpact": "HIGH"}, "impactScore": 5.9, "exploitabilityScore": 3.9}]}, "published": "2026-02-02T23:16:06.700", "references": [{"url": "https://github.com/vllm-project/vllm/pull/31987", "tags": ["Issue Tracking", "Patch"], "source": "security-advisories@github.com"}, {"url": "https://github.com/vllm-project/vllm/pull/32319", "tags": ["Issue Tracking", "Patch"], "source": "security-advisories@github.com"}, {"url": "https://github.com/vllm-project/vllm/releases/tag/v0.14.1", "tags": ["Release Notes"], "source": "security-advisories@github.com"}, {"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-4r2x-xpjr-7cvv", "tags": ["Vendor Advisory"], "source": "security-advisories@github.com"}], "vulnStatus": "Analyzed", "weaknesses": [{"type": "Primary", "source": "security-advisories@github.com", "description": [{"lang": "en", "value": "CWE-532"}]}], "descriptions": [{"lang": "en", "value": "vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1."}, {"lang": "es", "value": "vLLM es un motor de inferencia y servicio para modelos de lenguaje grandes (LLM). Desde la versi\u00f3n 0.8.3 hasta antes de la 0.14.1, cuando se env\u00eda una imagen inv\u00e1lida al endpoint multimodal de vLLM, PIL lanza un error. vLLM devuelve este error al cliente, filtrando una direcci\u00f3n de mont\u00edculo. Con esta fuga, reducimos ASLR de 4 mil millones de intentos a ~8 intentos. Esta vulnerabilidad puede encadenarse a un desbordamiento de mont\u00edculo con el decodificador JPEG2000 en OpenCV/FFmpeg para lograr ejecuci\u00f3n remota de c\u00f3digo. Esta vulnerabilidad est\u00e1 corregida en la versi\u00f3n 0.14.1."}], "lastModified": "2026-02-23T18:19:12.450", "configurations": [{"nodes": [{"negate": false, "cpeMatch": [{"criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*", "vulnerable": true, "matchCriteriaId": "6EBA2094-B756-41D7-A899-8C028DF95FF4", "versionEndExcluding": "0.14.1", "versionStartIncluding": "0.8.3"}], "operator": "OR"}]}], "sourceIdentifier": "security-advisories@github.com"}