Finding and attribution
Media acquisition can consume memory or bandwidth before size/item limits are applied. The advisory groups four paths under one boundary issue; authentication and exploitability depend on entry point and configuration, with an opt-in Rust tokenize route separately described.
Product: vLLM. AI system: GPT-5.5-Cyber. Team / organization: OpenAI; Trail of Bits.
CVE identifiers: Not established in the cited evidence.
AI and human contribution (source account): The project advisory credits GPT-5.5-Cyber through Patch the Planet, a Trail of Bits and OpenAI effort.
Human involvement is only described where the source account establishes it; unknown involvement does not mean none. Vendor confirmation of a flaw does not confirm every claim about AI contribution.
Discovery context: The cited source documents public disclosure; it does not establish whether the issue was previously known to defenders or exploited before disclosure.
Remediation status: fix documented. The advisory lists versions through 0.29.0 affected and versions >0.29.0 patched, while detailed reproduction confirms 0.25.1. Verify the lower bound and entry-point applicability before upgrading.
A source-documented fix does not establish installed patches or current safety. Source checking does not independently reproduce the finding or establish observed exploitation.
Disclosed 12 Sep 2026. AI attribution dated 12 Sep 2026. Sources checked 16 Nov 2026; these are separate dates, not a live installation check.
Recorded CVSS: Not available · no assessment established here. Documented technical effects: denial of service.
Severity, consequences and exploitation evidence
Case grouping uses the highest recorded CVSS score, with its version-specific severity band. It is descriptive, not a combined or editorial risk index. All recorded assessments remain below; scores from different versions and sources may disagree. CVSS specifications define version-specific scoring and severity bands.
No source-backed CVSS assessment is established in this catalog.
Case-level evidence: denial of service.
Prerequisites / affected scope: Media resource use can occur before size/item limits; exploitability and authentication conditions vary by entry point and configuration..
Exploitation evidence: The cited primary source does not report observed exploitation.. Impact source · checked 16 Nov 2026.
Documented fix: The advisory lists versions through 0.29.0 affected and versions >0.29.0 patched, while detailed reproduction confirms 0.25.1. Verify the lower bound and entry-point applicability before upgrading.