llama.cpp b10398 before b11301 Integer Overflow in gguf_init_from_reader
- Published
- Last modified
- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean? →
- published
- Whether CISA lists it in its Known Exploited Vulnerabilities catalog: proof that attackers are using it.What do these mean? →
- Not listed in CISA KEV
- Fix
- Fixed in b11301, 0.6.0
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
Summary
The llama.cpp inference library, used to run large language models from GGUF model files, is affected. An attacker who supplies a specially crafted GGUF file can trigger an integer overflow in tensor size checks, which may lead to out-of-bounds memory access and potentially crash or compromise the process loading the file.
Summarized by anthropic/claude-haiku-5.5. Verify details against the original sources.
Read the original description
llama.cpp b10398 before b11301 contains an integer overflow vulnerability in gguf_init_from_reader that allows attackers to bypass tensor element-count validation by supplying shapes whose product wraps to zero. Attackers can craft GGUF files with enormous tensor extents that overflow ggml_nbytes() and strides, causing out-of-bounds memory access when buffers are sized from wrapped byte counts.
Timeline
- CVE published
- First seen by cve.now
Affected versions
- ggml-org llama.cpp
From b10398 · fixed in b11301; From 0.1.0 · fixed in 0.6.0