PatchTheCVECVE-2026-100654: vLLM before 0.29.0 accepts user-controlled stop_token_ids on the OpenAI-compatible POST /v1/completions and PO
CVE-2026-100654 is a medium-severity flaw (CVSS 6.5) affecting vllm. Below: why it ranks where it does, which versions are affected, what to do, and the exact commands.
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 a
Why this priority
- No elevated risk signals yet
A fixed release or distro patch exists.
Risk facts
| CVSS score | 6.5 (Medium) |
| Attack vector | Network (remote) |
| Attack complexity | Low |
| Privileges needed | Low |
| User interaction | None |
| EPSS exploit chance | 0.3% (top 76%) |
| CISA exploited list | No |
| Weakness | CWE-129 |
| Published | 2026-09-26 |
| Last modified | 2026-10-07 |
Last updated 2026-10-07 · Sources: CVE Program, NVD, CISA KEV, EPSS and vendor advisories · How we score
Am I affected?
Timeline
- Published CVE record published
- Updated Record last modified
Scores and fixes here are computed automatically from those and other public sources. Confirm with the vendor advisory before changing production systems.
Quick answers
How do I fix CVE-2026-100654?
vllm: upgrade to 0.29.0. Exact commands for each OS are on this page.
Is CVE-2026-100654 being exploited?
CVE-2026-100654 is not on the CISA Known Exploited Vulnerabilities list at the time of the last update. Check again, as this can change.
What does CVE-2026-100654 affect?
CVE-2026-100654 affects vllm. Severity is medium (CVSS 6.5). 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 voca
How urgent is CVE-2026-100654?
Priority 29/100. No elevated risk signals yet.
Software pages
OpenAI and LLM tools vulnerabilities
Related CVEs
- CVE-2026-5497 HIGH 7.5 UPGRADE vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to u
- CVE-2026-48746 CRITICAL 9.1 UPGRADE vLLM is an inference and serving engine for large language models (LLMs)
- CVE-2026-54232 HIGH 8.8 UPGRADE vLLM is an inference and serving engine for large language models (LLMs)
- CVE-2026-41523 HIGH 7.5 UPGRADE vLLM is an inference and serving engine for large language models (LLMs)
- CVE-2026-54236 MEDIUM 5.3 UPGRADE vLLM is an inference and serving engine for large language models (LLMs)