vllm-project — vllm
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.
Available for technologies you monitor inside Seraph CTI.
Surfaced automatically when a documented actor is linked to this CVE.
Shown when this vendor/product is relevant to your industry.
Campaign and indicator context, where Seraph analysts have published it.
Declare your technology stack inside Seraph CTI and get notified the moment a vulnerability matches it.
