Vulnerability in ggml-org llama.cpp

[ Why it matters ]Priority signals

• Public proof-of-concept exploit exists (CISA SSVC)

[ Description ]Official 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.

[ Fixes ]Fixes and patches

No fix or interim patch is linked from our sources yet. Check the vendor references below and limit exposure of affected systems until a fix is available.

[ References ]References

• github.com/ggml-org/llama.cpp/blob/v0.5.0/ggml/src/gguf.cpp
• github.com/ggml-org/llama.cpp/commit/2149c00f4442dc59302e134a02e4c99d5f7ed9fc
• github.com/ggml-org/llama.cpp/issues/29383
• github.com/ggml-org/llama.cpp/pull/29384
• github.com/ggml-org/llama.cpp/releases/tag/b11301
• www.vulncheck.com/advisories/llama-cpp-b10398-before-b11301-integer-overflow-in-gguf-in…