Vulnerability Description
vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not that each token id is within the model vocabulary/logits range. When min_tokens > 0, the stop token ids are used as logits indices to suppress stop tokens, so an out-of-range id reaches a CUDA indexing operation (index_put_) and triggers a device-side assertion. An authenticated API user can send a single malformed completion request that returns 500 Internal Server Error and puts EngineCore into a fatal state, causing subsequent requests to fail until the service is restarted (denial of service).
CVSS Score
MEDIUM
Affected Products
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | < 0.29.0 |
Related Weaknesses (CWE)
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-v5gm-qgmv-gc6cMitigationVendor Advisory
- https://www.vulncheck.com/advisories/vllm-before-0.29.0-denial-of-service-via-ouThird Party Advisory
FAQ
What is CVE-2026-100654?
CVE-2026-100654 is a vulnerability with a CVSS score of 6.5 (MEDIUM). vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not tha...
How severe is CVE-2026-100654?
CVE-2026-100654 has been rated MEDIUM with a CVSS base score of 6.5/10. Review the CVSS metrics above for detailed severity breakdown.
Is there a patch for CVE-2026-100654?
Check the references section above for vendor advisories and patch information. Affected products include: Vllm Vllm.