CVE-2026-22773
- Source
- https://nvd.nist.gov/vuln/detail/CVE-2026-22773
- Import Source
- https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-22773.json
- JSON Data
- https://api.osv.dev/v1/vulns/CVE-2026-22773
- Aliases
- Published
- 2026-01-10T06:39:02.276Z
- Modified
- 2026-01-15T05:56:06.485160Z
- Severity
-
- 6.5 (Medium) CVSS_V3 - CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H CVSS Calculator
- Summary
- vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions
- Details
-
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.
- Database specific
{ "cwe_ids": [ "CWE-770" ], "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/22xxx/CVE-2026-22773.json", "cna_assigner": "GitHub_M" }- References
Affected packages
Git / github.com/vllm-project/vllm
Affected versions
v0.*
v0.10.0
v0.10.0rc1
v0.10.0rc2
v0.10.1rc1
v0.10.2rc1
v0.10.2rc2
v0.11.0rc1
v0.11.1
v0.11.1rc0
v0.11.1rc1
v0.11.1rc2
v0.11.1rc3
v0.11.1rc4
v0.11.1rc5
v0.11.1rc6
v0.6.4
v0.6.4.post1
v0.6.5
v0.6.6
v0.6.6.post1
v0.7.0
v0.7.1
v0.7.2
v0.7.3
v0.8.0rc1
v0.8.0rc2
v0.8.1
v0.8.2
v0.8.3rc1
v0.8.4
v0.9.0
v0.9.1
v0.9.1rc1
v0.9.1rc2
v0.9.2rc1