MEDIUMCVSS v3.1 5.5Vulnerability
CVE-2025-71439
llama.cpp before b5649 contains a heap out-of-bounds read vulnerability in llama_vocab::impl::print_info() that allows attackers to crash model loading via oversized special token IDs. Attackers can supply a crafted GGUF model with a short tokenizer.ggml.tokens array, causing llama-cli or llama-server to read past id_to_token and crash.
Properties
- Severity
- MEDIUM
- Nvd fetched at
- 2026-10-27T03:23:14.679434+00:00
- Cvss severity
- MEDIUM
- Cvss score
- 5.5
- Retrieved at
- 2026-10-27T03:24:29+00:00
- Score
- 5.5
- Cvss version
- 3.1
- Last source
- NVD
- Cve id
- CVE-2025-71439
- Cvss vector
- CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
- First seen by Signal
- 2026-10-27T03:16:31+00:00
- Vector
- CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
- Published at
- 2026-10-27T01:19:03.470
- Last modified
- 2026-10-27T01:19:03.470
Related Entities (1)
HAS_WEAKNESS (1)
Explore deeper with Ninja Signal's threat intelligence graph