CVE-2025-71439 json
llama.cpp before b5649 Heap Out-of-Bounds Read via llama_vocab::impl::print_info
published 2026-10-27 13 d ago modified 2026-11-05 3 d ago also EUVD-2025-211201
- cvss
- 5.5
- version 3.1, CNA
- cisa kev
- Not listed
- not in CISA's catalog
- epss
- 0.1%
- EPSS 0.1% chance of exploitation in 30 days, percentile 1%
- public exploits
- 0
- none known
Summary
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.
Patch / fix
- ggml-org llama.cpp b5649
- llama.cpp Debian 14 5713+dfsg-1
debian advisories DEBIAN-CVE-2025-71439
- github.com/ggml-org/llama.cpp@c33fe8b commit
- github.com/ggml-org/llama.cpp#14145 pull request
- github.com/ggml-org/llama.cpp/releases/tag/b5649 release
CISA SSVC
- Exploitation
- Proof of concept is public
- Automatable
- no
- Technical impact
- partial
- Assessed
- 4 d ago
Exploit prediction
0.12% chance of exploitation in the next 30 days, higher than 1% of all scored CVEs (FIRST EPSS, 2026-11-08).
Exploits and checks
- github.com https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-g4cc-763q-h9h6 referenced PoC, tagged "exploit" by a source, unverified
CVSS scores
| assigned by | version | score | vector |
|---|---|---|---|
| CNA | 4.0 | 6.8 | CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N |
| CNA | 3.1 | 5.5 | CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H |
| CNA | 4.0 | 6.8 | CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X |
Affected products
- ggml-org llama.cpp
Classification
- Weakness
-
- CWE-125 Out-of-bounds Read
References
-
github.com
/ggml-org/llama.cpp/pull/14145
patch, issue-tracking
-
github.com
/ggml-org/llama.cpp/commit/c33fe8b8c4427202706b1434e9fc8ab5752c9cac
patch
-
github.com
/ggml-org/llama.cpp/security/advisories/GHSA-g4cc-763q-h9h6
vendor-advisory, exploit
-
github.com
/ggml-org/llama.cpp/blob/b5648/src/llama-vocab.cpp#L2771
technical-description
-
github.com
/ggml-org/llama.cpp/releases/tag/b5649
release-notes
-
vulncheck.com
/advisories/llama-cpp-before-b5649-heap-out-of-bounds-read-via-llama-vocab-impl-print-info
third-party-advisory
Sources
- CVE List CVE-2025-71439
- NVD CVE-2025-71439
Fencer Vulns keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.
CPEs and fixed versions use Anchore's correction of NVD's data: Added CPE configurations because not yet analyzed by NVD.