{"cve":"CVE-2026-22773","nvd":{"published":"2026-01-10T07:16:03.527","modified":"2026-06-17T10:20:24.027","status":"Analyzed","description":"vLLM is an inference and serving engine for large language models (LLMs). In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimension mismatch that results in an unhandled runtime error, leading to complete server termination. This issue has been patched in version 0.12.0.","cvss_v4":null,"cvss_v3":{"version":"3.1","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","score":7.5},"cwes":["CWE-770"],"references":["https://github.com/vllm-project/vllm/security/advisories/GHSA-grg2-63fw-f2qr"],"products":["vllm:vllm"],"source":"nvd","synced_at":"2026-10-19T22:19:41.481987139+00:00"},"epss":{"cve":"CVE-2026-22773","score":0.00442,"percentile":0.38414,"date":"2026-10-31"},"kev":null,"osv":[{"id":"GHSA-grg2-63fw-f2qr","url":"https://osv.dev/vulnerability/GHSA-grg2-63fw-f2qr","summary":"vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions","modified":"2026-09-10T03:50:33.155799536Z","packages":[{"ecosystem":"PyPI","name":"vllm","purl":"pkg:pypi/vllm","ranges":[">=0.6.4 <0.12.0"],"fixed":["0.12.0"]}]},{"id":"PYSEC-2026-143","url":"https://osv.dev/vulnerability/PYSEC-2026-143","summary":"","modified":"2026-05-20T09:19:21.486475Z","packages":[{"ecosystem":"PyPI","name":"vllm","purl":"pkg:pypi/vllm","ranges":[">=0.6.4 <0.12.0"],"fixed":["0.12.0"]}]}],"attribution":["This product uses the NVD API but is not endorsed or certified by the NVD.","EPSS scores: FIRST Exploit Prediction Scoring System, https://www.first.org/epss","Known exploited vulnerabilities: CISA Known Exploited Vulnerabilities Catalog, https://www.cisa.gov/known-exploited-vulnerabilities-catalog","Package advisories: OSV (https://osv.dev), from the GitHub Advisory Database, PyPA, Go, RustSec and other ecosystem databases, under CC-BY 4.0 or CC0","CVE records: the CVE Program's CVE List (https://www.cve.org), under its terms of use (https://www.cve.org/Legal/TermsOfUse). CVE is a registered trademark of The MITRE Corporation; this product is not affiliated with the CVE Program."]}