---
title: "CVE-2026-116816: llama.cpp - Out of Bounds Read/Write"
description: "llama.cpp before b10261 contains a heap out-of-bounds write caused by improper handling of BYTE-typed token text in llm_tokenizer_plamo2::build, letting attackers corrupt heap memory via crafted GGUF models, exploit requires crafted model input."
url: https://panop.io/threat-heartbeat/CVE-2026-116816/
---

# CVE-2026-116816 llama.cpp - Out of Bounds Read/Write

llama.cpp before b10261 contains a heap out-of-bounds write caused by improper handling of BYTE-typed token text in llm\_tokenizer\_plamo2::build, letting attackers corrupt heap memory via crafted GGUF models, exploit requires crafted model input.

CVSS base score 7.8 /10

High

CISA KEV

Not listed

Public exploit

None known

Authentication

Not required

[View the NVD record ↗](https://nvd.nist.gov/vuln/detail/CVE-2026-116816)

## Details

Type

Out of Bounds Read/Write

Impact

Remote Code Execution, Denial Of Service

CVSS vector

CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Published

27 October 2026

Updated

27 October 2026

Weakness

[CWE-129](https://cwe.mitre.org/data/definitions/129.html)

## Vendor fix

Update to version b10261 or later.

Patching closes this CVE. It does not tell you which of your systems run the affected version, or what else is exposed alongside them.

## References

- [https://nvd.nist.gov/vuln/detail/CVE-2026-116816](https://nvd.nist.gov/vuln/detail/CVE-2026-116816)
- [https://github.com/ggml-org/llama.cpp/pull/25969](https://github.com/ggml-org/llama.cpp/pull/25969)
- [https://github.com/ggml-org/llama.cpp/pull/26511](https://github.com/ggml-org/llama.cpp/pull/26511)
- [https://github.com/ggml-org/llama.cpp/releases/tag/b10261](https://github.com/ggml-org/llama.cpp/releases/tag/b10261)
- [https://www.vulncheck.com/advisories/llama-cpp-b5899-before-b10261-heap-out-of-bounds-write-via-plamo2-tokenizer](https://www.vulncheck.com/advisories/llama-cpp-b5899-before-b10261-heap-out-of-bounds-write-via-plamo2-tokenizer)
- [https://github.com/ggml-org/llama.cpp/blob/b10259/src/llama-vocab.cpp#L1376-L1377](https://github.com/ggml-org/llama.cpp/blob/b10259/src/llama-vocab.cpp#L1376-L1377)
- [https://github.com/ggml-org/llama.cpp/commit/cf06ad7dfe79e7bf2934150d4fefe1ce8db05490](https://github.com/ggml-org/llama.cpp/commit/cf06ad7dfe79e7bf2934150d4fefe1ce8db05490)

[← All CVEs](https://panop.io/threat-heartbeat/)

## What this record can't tell you

- Whether you run the affected software, and where
- Whether an attacker can actually reach and exploit it
- The misconfigurations that never get a CVE number
- What it would cost the business, ranked into what to fix first

Panop answers these inside your environment, for this CVE and every other finding.

[Book a demo](https://panop.io/demo)
