Vulnerability Description
vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.
CVSS Score
MEDIUM
Affected Products
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | >= 0.22.0, < 0.24.0 |
Related Weaknesses (CWE)
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-qff2-492f-9fm4ExploitMitigationVendor Advisory
- https://www.vulncheck.com/advisories/vllm-0.22.0-through-0.23.0-denial-of-servicThird Party Advisory
FAQ
What is CVE-2026-100652?
CVE-2026-100652 is a vulnerability with a CVSS score of 5.9 (MEDIUM). vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. A...
How severe is CVE-2026-100652?
CVE-2026-100652 has been rated MEDIUM with a CVSS base score of 5.9/10. Review the CVSS metrics above for detailed severity breakdown.
Is there a patch for CVE-2026-100652?
Check the references section above for vendor advisories and patch information. Affected products include: Vllm Vllm.