Vulnerabilidad · Publicado 21/07/2026 · Actualizado 01/08/2026
Ollama (HEAD f0078ae) contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata array counts that are used as allocation sizes without validation against remaining file size. Attackers …
Ollama (HEAD f0078ae) contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata array counts that are used as allocation sizes without validation against remaining file size. Attackers can upload a sub-1KB crafted GGUF file via the blob upload and model create or pull API endpoints to trigger unrecoverable Go runtime out-of-memory fatal errors or makeslice panics that bypass recovery middleware and crash the entire server process.
Score: 7.5/10 — Severidad: HIGH — Estado NIST: Awaiting Analysis
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
CWE-789
Publicado en NIST NVD.