CVE-2026-93841: vLLM through 0.29.0 Adjacent-Request Sampler State Corruption via Unvalidated Prompt Token IDs

Published Sep 18, 2026
·
Updated

vLLM through 0.29.0 contains a memory corruption vulnerability in the Triton bincountkernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size. Attackers can submit multimodal audio requests with tokens equal to vocabulary size, causing out-of-bounds writes that corrupt concurrent requests' sampler state and alter repetition penalty behavior.

Affected Software

1 affected component
vllm vllm<=0.29.0

Event History

Sep 18, 2026
CVE Published
via MITRE·07:06 PM
Data Sourced
via MITRE·07:06 PM
DescriptionSeverityWeakness

Frequently Asked Questions

1

Does exploiting this issue require authentication or user interaction?

No. The CVSS vector indicates no privileges are required and no user interaction is needed, although exploitation has high attack complexity.

2

Does the reported impact include data disclosure or denial of service?

No confidentiality or availability impact is reported in the CVSS vector. The reported impact is limited to low-integrity changes affecting repetition-penalty behavior in concurrent requests.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203