{"action":"Upgrade llama.cpp to 7965+dfsg-1 or later.","affected":[{"fixed":"7965+dfsg-1","introduced":"0"}],"component":"llama.cpp","cve_ids":[],"ecosystem":"Debian","ecosystem_qualifier":"Debian:14","exploited":false,"fixed_versions":["7965+dfsg-1"],"id":"tcs_01M5NRQHBTZAHPKAQ48PQSVATH","modified_at":"2026-10-01T03:02:37.341123257Z","published_at":"2026-03-24T01:17:01.870Z","references":["https://osv.dev/vulnerability/DEBIAN-CVE-2026-33298","https://security-tracker.debian.org/tracker/CVE-2026-33298"],"severity":{"label":"high","score":7.8},"sources":[{"source":"osv","source_id":"DEBIAN-CVE-2026-33298","url":"https://osv.dev/vulnerability/DEBIAN-CVE-2026-33298"}],"summary":"llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a G"}