CVE-2026-115916
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact multiple-of-64 vocabulary, VocabParallelEmbedding.forward does not mask that value before F.embedding uses it, causing a CUDA device-side assertion and terminating the engine. This issue is fixed in version 0.32.0.
Am I affected?
Sign in and add the software you use to your watchlist: we check it against every new vulnerability.
Show me how to fix itTimeline
- Published by NVD
- Last updated by NVD
Affected products
| Product | Versions |
|---|---|
| Vllm Vllm | up to before 0.32.0 |
Sources: NVD and CISA known exploited vulnerabilities catalogue.