CVE-2026-5497
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.
Does this matter?
High impact if exploited, but EPSS currently rates exploitation as unlikely (0.54%). Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.
Description
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.
- CVSS 3.1
- 7.5 HIGHCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
- EPSS
- 0.54% probability · 44th percentile
- CISA KEV
- Not listed
- Weakness
- CWE-400, CWE-770
- Affected
- vllm/vllm
- Source
- security@huntr.dev
References
- https://github.com/vllm-project/vllm/commit/58ee61422169ce17e08248f8efa1e9df434fe395Patch
- https://huntr.com/bounties/7bd92629-b396-4449-8f88-6c0092530eb4Exploit, Third Party Advisory
- https://access.redhat.com/security/cve/CVE-2026-5497
- https://bugzilla.redhat.com/show_bug.cgi?id=2487813
- https://huntr.com/bounties/7bd92629-b396-4449-8f88-6c0092530eb4Exploit, Third Party Advisory
- https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-5497.json
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.