Out-of-bounds Read in Ggml-Org Llama.cpp

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.

Vulnerability class: Buffer Overflow

Published · last modified .

CVSS v3 metric

CVSS v3 base score 5.5 (Medium). Vector: CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H.

EPSS exploit prediction

EPSS: 0.001 (1.9th percentile), scored .

Very low probability of exploitation in the next 30 days; routine patching cadence is appropriate. 2nd percentile — 1.9% of CVEs in the catalogue have a lower EPSS than this one. How to read EPSS.

The 30-day trend appears once there are two daily scores.

Affected products

Weakness classification (CWE)

References

Frequently asked questions

What is CVE-2025-71439?
CVE-2025-71439 is a medium-severity vulnerability in Ggml-Org Llama.cpp, classified under Out-of-bounds Read. CVSS score: 5.5/10. Published 2026-10-27.
How severe is CVE-2025-71439?
Medium severity. CVSS v3 base score is 5.5 out of 10.