PYSEC-2026-2302
Details
vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.load_base64()` method. When processing `video/jpeg` data URLs, the method splits the base64 data string on commas to extract individual JPEG frames without enforcing a frame count limit. An attacker can exploit this by crafting a single API request containing thousands of comma-separated base64-encoded JPEG frames in a data URL, causing the server to decode all frames into memory and crash due to excessive memory consumption. This vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.
Are you affected?
Enter the version of the package you're using.
Affected packages
References
- https://access.redhat.com/security/cve/CVE-2026-5497 [WEB]
- https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-5497.json [WEB]
- https://bugzilla.redhat.com/show_bug.cgi?id=2487813 [REPORT]
- https://github.com/vllm-project/vllm/commit/58ee61422169ce17e08248f8efa1e9df434fe395 [FIX]
- https://huntr.com/bounties/7bd92629-b396-4449-8f88-6c0092530eb4 [EVIDENCE]