CVE-2026-116840

EUVD-2026-110571
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. Attackers can supply a crafted GGUF file whose tensor size pads to zero, causing an undersized data blob and out-of-bounds reads by consumers.
ProviderTypeBase ScoreAtk. VectorAtk. ComplexityPriv. RequiredVector
VulnCheckCNA
6.1 MEDIUM
LOCAL
LOW
NONE
CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:L/I:N/A:H
Base Score
CVSS 3.x
EPSS Score
Percentile: 2.3%
Early Detection
Affected products identified ahead of NVD analysis through intelligence sources (CNA/ADP records, vendor advisories, Enginsight analysis).
VendorProductVersionSource
ggmlllama.cpp
0.1.0 ≤
𝑥
< 0.6.0
CNA
Debian logo
Debian Releases
Debian Product
Codename
llama.cpp
forky
1:0.7.0+dfsg-2
fixed
sid
1:0.7.0+dfsg-2
fixed
Common Weakness Enumeration
References
https://github.com/ggml-org/llama.cpp/blob/v0.5.0/ggml/src/gguf.cpp#L794-L795
https://github.com/ggml-org/llama.cpp/commit/a6ea155d3d38b3f6f43d0c0dc29c2d592413ef44
https://github.com/ggml-org/llama.cpp/issues/26978
https://github.com/ggml-org/llama.cpp/pull/26979
https://github.com/ggml-org/llama.cpp/releases/tag/b11275
https://www.vulncheck.com/advisories/llama-cpp-b5854-before-b11275-integer-overflow-in-gguf-init-from-reader