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.
CVSS Details
- CVSS 4.0 Base Score: 6.3 (MEDIUM)
- CVSS 4.0 Vector: (CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X)
- CVSS 3.1 Base Score: 3.7
- CVSS 3.1 Vector: (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N)
Prioritise with Active Threat Intelligence
With curated Threat Intelligence, you can see which vulnerabilities truly put you at risk, prioritize what matters most, and act before attackers do.
Explore Intelligence Hub