CVE-2026-33298 llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing
- Published
- Updated
- Exploit status
- No public exploit known
- CISA SSVC
- Proof of concept public · automatable: no · technical impact: total
- Technology
- UbuntuDebian
Description
llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the ggml_nbytes function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes ggml_nbytes to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix.
Upgrade path
5 packages: 2 fixed, 2 with no fix yet.
| Package | Affected | Fixed in | Status |
|---|---|---|---|
| ubuntu/llama.cpp(Ubuntu 26.04) | All versions | No fix listed | No fix yet |
| ubuntu/llama.cpp(Ubuntu 26.10) | All versions | No fix listed | No fix yet |
| debian/llama.cpp(Debian 14) | < 7965+dfsg-1; = 5151+dfsg-1~exp2; = 5151+dfsg-1~exp3; = 5318+dfsg-1 +19 more | 7965+dfsg-1 | Fixed |
| ggml llama.cpp | < b7824 | b7824 | Fixed |
| ggml-org llama.cpp | = < b7824 | No fix listed | Versions listed |
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| ggml-org llama.cpp | = < b7824 | No fix listed |
| debian/llama.cpp(Debian:14) | < 7965+dfsg-1 | 7965+dfsg-1 |
| debian/llama.cpp(Debian:14) | = 5151+dfsg-1~exp2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5151+dfsg-1~exp3 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5318+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5318+dfsg-2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5713+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5760+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5760+dfsg-2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5760+dfsg-3 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5760+dfsg-4 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-3 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-3~exp1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-3~exp2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-3~exp3 | No fix listed |
| debian/llama.cpp(Debian:14) | = 5882+dfsg-4 | No fix listed |
| debian/llama.cpp(Debian:14) | = 6641+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 6641+dfsg-2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 6641+dfsg-3 | No fix listed |
| debian/llama.cpp(Debian:14) | = 7593+dfsg-1 | No fix listed |
| debian/llama.cpp(Debian:14) | = 7593+dfsg-2 | No fix listed |
| debian/llama.cpp(Debian:14) | = 7593+dfsg-3 | No fix listed |
| ubuntu/llama.cpp(Ubuntu:26.04) | All versions | No fix listed |
Show 2 more
| Product | Affected versions | Fixed in |
|---|---|---|
| ubuntu/llama.cpp(Ubuntu:26.10) | All versions | No fix listed |
| ggml llama.cpp | < b7824 | b7824 |
References
- https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-96jg-mvhq-q7q7 exploit vendor-advisory
- https://github.com/ggml-org/llama.cpp/releases/tag/b7824 release-notes
- Debian Security Tracker CVE-2026-33298 vendor-advisory
- Ubuntu CVE tracker vendor-advisory
- www.cve.org