Vulnerability · Published Jan 7, 2026 · Assigned by GitHub_M

CVE-2026-21869llama.cpp has Out-of-bounds Write in llama-server

High 8.8

In plain English

A high-severity out-of-bounds write in llama.cpp from ggml-org. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.5% (higher than 45% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.

Severity (CVSS 3.1)8.8High

Exploit odds (30 days)0.5%top 55.1% of all CVEs

Exploited in the wildNo reportsCISA SSVC: poc

Attack toolingNoneno public tooling known

Official description

llama.cpp is an inference of several LLM models in C/C++. In commits 55d4206c8 and prior, the n_discard parameter is parsed directly from JSON input in the llama.cpp server's completion endpoints without validation to ensure it's non-negative. When a negative value is supplied and the context fills up, llama_memory_seq_rm/add receives a reversed range and negative offset, causing out-of-bounds memory writes in the token evaluation loop. This deterministic memory corruption can crash the process or enable remote code execution (RCE). There is no fix at the time of publication.

Affected packages & fixed versions

Ecosystem	Package	Fixed in	Advisory
Debian:14	llama.cpp	7593+dfsg-1	DEBIAN-CVE-2026-21869
Ubuntu:25.10	llama.cpp	no fix listed	UBUNTU-CVE-2026-21869
Ubuntu:26.04:LTS	llama.cpp	no fix listed	UBUNTU-CVE-2026-21869
Ubuntu:26.10	llama.cpp	no fix listed	UBUNTU-CVE-2026-21869

Affected products (as reported)

Vendor	Product	Versions
ggml-org	llama.cpp	<= 55d4206c8

Weakness: CWE-787 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

References

https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-8947-pfff-2f3c x_refsource_CONFIRM