Code execution via wrapped GGUF tensor size
Published Mar 24, 2026 · Updated Mar 24, 2026
Integer overflow in llama.cpp before b7824 allows attackers to execute code or crash applications by supplying a malicious GGUF model. The ggml_nbytes function multiplies attacker-controlled tensor dimensions and strides without detecting size_t wrap, so the loader allocates a small heap buffer for an enormous tensor. Exploitation requires an application to load the untrusted model; subsequent tensor access overruns the allocation and can corrupt memory or terminate the process.
Summary
What happened
Integer overflow in llama.cpp before b7824 allows attackers to execute code or crash applications by supplying a malicious GGUF model. The ggml_nbytes function multiplies attacker-controlled tensor dimensions and strides without detecting size_t wrap, so the loader allocates a small heap buffer for an enormous tensor. Exploitation requires an application to load the untrusted model; subsequent tensor access overruns the allocation and can corrupt memory or terminate the process.
The record
- CVE
- CVE-2026-33298
- Published
- Mar 24, 2026
- Updated
- Mar 24, 2026
- Vendor
- Georgi Gerganov
- Product
- llama.cpp
- Classifications
- CWE-122, CWE-190, T1204.002, T1203
- Attack vector
- local
- Privileges
- unauthenticated
Timeline
How it unfolded
- Mar 24, 2026CVE publishedPublication date reported by the CVE source.
- Mar 24, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< b7824
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- Malicious GGUF tensor-dimension proof of conceptproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo