Tools AI Risk Radar ai-incident-0157
Incident record
vLLM fixes crashes after rejected media requests
- Severity
- Medium
- Status
- Patched
- Type
- Infra Vuln
- Target
- vllm
- Actor
- researcher
- CVE
- CVE-2026-105753
What happened
A rejected request could leave two media caches out of step. A later request reusing the same media identifier could then trigger an assertion and stop the shared inference engine.
vllm: affected < 0.28.0; fixed in 0.28.0. The date shown is the maintainer’s original disclosure date. Its advisory assigns 6.5 under CVSS 3.1.
Sources
- Primary disclosuregithub.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5…
One record from the AI Risk Radar, maintained by the Security Desk.Data: CSV · JSON ·RSS · CC BY 4.0 with attribution to YFarmX.Tracker updated · 13 October 2026