LOWCVSS 3.7
vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation, causing LogitBiasState to corrupt GPU logits state and allow concurrent requests to sample tokens outside their allowlists.
Published 2026-09-18 · last modified 2026-09-18
← Get alerted the moment a CVE hits your gear — subscribe freeNcerio by BeyondNets · data from NVD, CISA KEV, EPSS.