Ggml-org llama.cpp
llama.cpp has a Heap Buffer Overflow via Integer Overflow in `mem_size` Calculation — Bypass of CVE-2025-53630 Fix
llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file - CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.
Affected versions and mitigation
From the CVE record and the vendor's references.Affected
- < b8146
Mitigation
The record names no fix. Check the vendor advisory below.
References
- Vendor advisory github.com/ggml-org/llama.cpp/security/advisor…
Timeline
Days relative to CVE publicationFirst public PoC 17 days after publication; not in CISA KEV. First PoC: ngtuonghung/CVE-2026-27940 (created 29 Mar 2026, found by us: backfilled).
PoC repositories
Found on GitHub.1 PoCs
PoCs 1
| Repository | Created | Stars | Assessment |
|---|---|---|---|
| ngtuonghung/CVE-2026-27940 | 29 Mar 2026 +17 d | 1 | PoC via PoC Index |
Only PoCs count toward the first-PoC date and the characteristics below. Do not run PoC code from unknown sources.