Loading vulnerability…
CVE-2026-27940 affects llama.cpp, a C/C++ inference implementation for multiple LLM models. Prior to commit b8146, the function gguf_init_from_file_impl() in gguf.cpp contains an integer overflow condition that results in an undersized heap allocation. A subsequent fread() then writes 528 or more bytes of attacker-controlled data beyond the allocated buffer boundary, causing a heap-based out-of-bounds write. The issue is described as a bypass of a similar flaw previously addressed as CVE-2025-53630, where the earlier fix did not cover all affected code paths. The vulnerability is fixed in b8146.
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
No valid public exploits. Mallory filtered out 1 candidate as fakes, detection scripts, or README-only repos.
All candidate exploits were filtered out by Mallory's validation.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
Vendor-confirmed product mapping. Mallory continuously reconciles this list against your asset inventory.
3 sources tracked across advisories and community write-ups. News coverage will land here when it surfaces.
No news coverage yet. Advisories and community discussion only.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.