CVE-2026-43632
CRITICAL
INFO
Priority: 0.04
BB Relevance:
68/100
Published: 2026-08-06
Modified: 2026-10-23
First seen: 2026-10-29
Status: active
Attack Surface
Network
No Auth
No Interaction
Conf. Impact
Integ. Impact
CVSS Vector Breakdown
AV:
Network
AC:
High
PR:
None
UI:
None
S:
X
AT:
Present
VC:
High
VI:
High
VA:
High
Description
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
References & Advisories (2)
Patch
Third Party Advisory
Patch
Third Party Advisory