CVE-2026-93841
Description
vLLM through 0.29.0 contains a memory corruption vulnerability in the Triton _bincount_kernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size. Attackers can submit multimodal audio requests with tokens equal to vocabulary size, causing out-of-bounds writes that corrupt concurrent requests' sampler state and alter repetition penalty behavior.
How to fix
- vllm0.30.0 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| vllm vllm | v ≤ 0.29.0 | — |
| vllmPyPI | v < 0.30.0 | 0.30.0 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.