Skip to content
OpenAdvisories
High · 7.8 CVE-2026-43628 EUVD-2026-54283GHSA-wvf4-v63v-mwm5

llama.cpp b3978–b9058 Integer Underflow via DRY Sampler

Published Aug 6, 2026 · Updated Oct 29, 2026 · Assigned by VulnCheck

Description

llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.

Affected products

VendorProductAffectedFixed
ggml-org llama.cpp b3978 – ≤ b9058 —

Fixed in Linux distributions

Distributions backport fixes, so the version that fixes this vulnerability differs per release — upgrade to at least the version shown. From the distributions' security trackers and advisories.

ReleasePackageFixed inAdvisory
Debian 14 llama.cpp No fix yet —
ReleasePackageFixed inAdvisory
Ubuntu 26.10 llama.cpp No fix yet —
Ubuntu 26.04 LTS llama.cpp No fix yet —

References