Medium severity · published Sep 18, 2026
CVE-2026-93840
vLLM before 0.29.0 Cross-Request Logits Corruption via allowed_token_ids
5.3
CVSS
0.2%
EPSS
No
On KEV
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.
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N
Attack vector
Network
Complexity
Low
Privileges
None
User interaction
None
Confidentiality
None
Integrity
Low
Availability
None
Scope
Unchanged
Exploitation
Is it being used?
Exploit prediction (EPSS)0.2% · higher than 16% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
VL
vllm · vllm-project
before 0.29.0 · source: vendor, NVD