CVE-2026-93840

Description

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.

Affected Platforms

Weakness Type

CWE-129

References

• https://github.com/vllm-project/vllm
• https://github.com/vllm-project/vllm/blob/v0.28.0/vllm/sampling_params.py#L881-L…
• https://github.com/vllm-project/vllm/blob/v0.28.0/vllm/v1/worker/gpu/sample/logi…
• https://github.com/vllm-project/vllm/commit/5b0e5b69ac1a3884a6479c9537789c95263c…
• https://github.com/vllm-project/vllm/pull/49080
• https://www.vulncheck.com/advisories/vllm-before-0.29.0-cross-request-logits-cor…