← Vulnerability feed

Vulnerability record · CVE-2026-43632 · published 6 August 2026

CVE-2026-43632: Ggml llama.cpp toctou race condition vulnerability

Ggml · Llama.Cpp

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.

9.2 CVSS 4.0 Critical EPSS 0.48% · top 61.3% CWE-367 · TOCTOU race conditionCWE-416 · Use after free
9.2CVSS 4.0 base score
0.48%EPSS exploitation probability, 30 days
NoNot in CISA KEV
1Affected product versions listed by NVD
2References
21 Sep 2026Last modified by NVD

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.

CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:

Affected products

1 vulnerable configurations from NVD's CPE data, grouped by vendor and product.

References

LinkTags
https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches PatchThird Party Advisory
https://www.vulncheck.com/advisories/llama-cpp-b7492-b9060-use-after-free-in-tokenization-endpoints PatchThird Party Advisory

Track CVE-2026-43632 inside VULONE

Watch it alongside the ransomware crews, C2 infrastructure and forum chatter that reference it, query it through the API and pull it into your SIEM over TAXII.

Start free Open in platform

Related vulnerabilities

Same products first, then exploited flaws of the same weakness class.

9.8CVE-2026-34159Ggml llama.cpp memory buffer overflow vulnerabilityllama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation …EPSS 1.2%9.8CVE-2026-21869Ggml llama.cpp out-of-bounds write vulnerabilityllama.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 …EPSS 0.52%9.8CVE-2024-42478Ggml llama.cpp out-of-bounds read vulnerabilityllama.cpp provides LLM inference in C/C++. The unsafe `data` pointer member in the `rpc_tensor` structure can cause arbitrary address reading. This v…EPSS 0.60%9.8CVE-2024-42479Ggml llama.cpp out-of-bounds write vulnerabilityllama.cpp provides LLM inference in C/C++. The unsafe `data` pointer member in the `rpc_tensor` structure can cause arbitrary address writing. This v…EPSS 2.6%9.8CVE-2024-23605Ggml llama.cpp integer overflow vulnerabilityA heap-based buffer overflow vulnerability exists in the GGUF library header.n_kv functionality of llama.cpp Commit 18c2e17. A specially crafted .ggu…EPSS 1.3%9.8CVE-2024-23496Ggml llama.cpp integer overflow vulnerabilityA heap-based buffer overflow vulnerability exists in the GGUF library gguf_fread_str functionality of llama.cpp Commit 18c2e17. A specially crafted .…EPSS 1.3%9.8CVE-2024-21802Ggml llama.cpp heap-based buffer overflow vulnerabilityA heap-based buffer overflow vulnerability exists in the GGUF library info->ne functionality of llama.cpp Commit 18c2e17. A specially crafted .ggu…EPSS 1.4%9.8CVE-2024-21825Ggml llama.cpp integer overflow vulnerabilityA heap-based buffer overflow vulnerability exists in the GGUF library GGUF_TYPE_ARRAY/GGUF_TYPE_STRING parsing functionality of llama.cpp Commit 18c2…EPSS 1.3%

Source: NIST National Vulnerability Database (record CVE-2026-43632), CISA KEV, FIRST EPSS (scores of 2026-09-26). This page is refreshed as NVD updates the record.