PatchSiren cyber security CVE debrief
CVE-2026-5497 vllm-project CVE debrief
CVE-2026-5497 is a HIGH severity vulnerability in vLLM, a library for large language models. 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. This method processes `video/jpeg` data URLs by splitting 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. The vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.
- Vendor
- vllm-project
- Product
- vllm-project/vllm
- CVSS
- HIGH 7.5
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-06-11
- Original CVE updated
- 2026-07-22
- Advisory published
- 2026-06-11
- Advisory updated
- 2026-07-22
Who should care
Users of vLLM library, particularly those using versions 0.8.0 and later, should be aware of this vulnerability and take steps to mitigate it.
Technical summary
The `VideoMediaIO.load_base64()` method in vLLM versions 0.8.0 and later does not limit the number of JPEG frames that can be extracted from a `video/jpeg` data URL. This allows an attacker to cause a Denial of Service (DoS) by providing a URL with thousands of comma-separated base64-encoded JPEG frames, leading to excessive memory consumption and a potential crash.
Defensive priority
HIGH
Recommended defensive actions
- Update to a patched version of vLLM, if available.
- Implement input validation and frame count limiting for `video/jpeg` data URLs.
- Monitor API usage for suspicious activity.
Evidence notes
Evidence for this CVE comes from the NVD and Huntr.
Sources and references
Verified primary and authoritative sources
-
CVE-2026-5497 CVE Program record
Publisher, destination, and source semantics verified
URL: https://www.cve.org/CVERecord?id=CVE-2026-5497
CVE Program - Official CVE Program record with source-provided CVE metadata.
-
CVE-2026-5497 NVD vulnerability detail
Publisher, destination, and source semantics verified
URL: https://nvd.nist.gov/vuln/detail/CVE-2026-5497
NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.
Supplemental references
-
Source reference
Unverified legacy reference
URL: https://github.com/vllm-project/vllm/commit/58ee61422169ce17e08248f8efa1e9df434fe395
-
Source reference
Unverified legacy reference
URL: https://huntr.com/bounties/7bd92629-b396-4449-8f88-6c0092530eb4
Methodology and review provenance
AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.