CVE-2026-100654
vLLM before 0.29.0 Denial of Service via out-of-range stop_token_ids
- State
- Published
- Severity
- 7.1 High
- Published
- Sep 26, 2026
- Updated
- Sep 30, 2026
- Reserved
- Sep 26, 2026
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).
Scores
| CVSS | Score | Vector | Source |
|---|---|---|---|
| 4.0 | 7.1 High | CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N | VulnCheck |
| 3.1 | 6.5 Medium | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | VulnCheck |
Weaknesses
- CWE-129 Improper Validation of Array Index
Affected Products
| Product | Versions |
|---|---|
vllmvllm-project |
|
References
- GitHub Security Advisory (GHSA-v5gm-qgmv-gc6c)vendor-advisory
- VulnCheck Advisory: vLLM before 0.29.0 Denial of Service via out-of-range stop_token_idsthird-party-advisory
Every record here comes from the CVE Program’s CVE List. © 1999–2026 The MITRE Corporation. CVE is a registered trademark of The MITRE Corporation. The records are reproduced under the CVE Program terms of use.