{"id":"CVE-2025-53630","published":"2025-07-10T19:32:45Z","modified":"2025-07-10T20:31:07Z","title":"Integer Overflow in GGUF Parser can lead to Heap Out-of-Bounds Read/Write in gguf","vendor":"ggml-org","product":"llama.cpp","tier":"tail","description":{"text":"llama.cpp is an inference of several LLM models in C/C++. Integer Overflow in the gguf_init_from_file_impl function in ggml/src/gguf.cpp can lead to Heap Out-of-Bounds Read/Write. This vulnerability is fixed in commit 26a48ad699d50b6268900062661bd22f3e792579.","provenance":"source","from":"CVE.org"},"cvss":{"score":8.9,"severity":"HIGH","version":"4.0","vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:P","provenance":"source","from":"CVE.org CNA"},"epss":{"score":0.00363,"percentile":0.30132,"provenance":"source"},"cwes":[{"id":"CWE-122","name":"Heap-based Buffer Overflow"},{"id":"CWE-680","name":"Integer Overflow to Buffer Overflow"}],"references":[{"url":"https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-vgg9-87g3-85w8","tags":["x_refsource_CONFIRM"]},{"url":"https://github.com/ggml-org/llama.cpp/commit/26a48ad699d50b6268900062661bd22f3e792579","tags":["x_refsource_MISC"]}],"priority":{"score":12.3,"factors":[{"name":"EPSS percentile","points":6.03},{"name":"CVSS severity","points":6.23}]},"vendorSlug":"ggml-org","affectedVersions":[{"vendor":"ggml-org","product":"llama.cpp","versions":["< 26a48ad699d50b6268900062661bd22f3e792579"]}],"euvd":{"id":"EUVD-2025-21070","exploited":false,"provenance":"source"},"lineage":{"familyId":"fam-ggml-org-llama-cpp-cve-2025-53630","children":[{"from":"CVE-2025-53630","to":"CVE-2026-116840","relation":"bypass","confidence":"high","kind":"lineage","evidence":[{"kind":"text","provenance":"source","relation":"bypass","quote":"llama.cpp b5854 before b11275 contains an integer overflow vulnerability in gguf_init_from_reader that bypasses the CVE-2025-53630 size check when padding near-SIZE_MAX tensor sizes.","from":"CVE.org","in":"CVE-2026-116840"},{"kind":"same-code","provenance":"machine","files":["ggml/src/gguf.cpp"],"commits":["https://github.com/ggml-org/llama.cpp/commit/26a48ad699d50b6268900062661bd22f3e792579","https://github.com/ggml-org/llama.cpp/commit/a6ea155d3d38b3f6f43d0c0dc29c2d592413ef44"],"basis":"shared function","functions":["gguf_context"]}]},{"from":"CVE-2025-53630","to":"CVE-2026-27940","relation":"related","confidence":"medium","kind":"lineage","evidence":[{"kind":"text","provenance":"source","relation":"related","quote":"This is a bypass of a similar bug in the same file - CVE-2025-53630, but the fix overlooked some areas.","from":"CVE.org","in":"CVE-2026-27940"},{"kind":"similar","provenance":"machine","score":0.4,"basis":"same product + CWE + description similarity"}]}]}}