Every CVE Search/ Get API key

CVE-2025-71439

llama.cpp before b5649 Heap Out-of-Bounds Read via llama_vocab::impl::print_info

6.8 Medium

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.

Attack profile

How easy it is to exploit

Attack vector
Local
Complexity
Low
Privileges needed
None
User interaction
Passive

What an attacker can affect

Confidentiality
None
Integrity
None
Availability
High

Exploit likelihood

0.12%Unchanged since Oct 27, 2026
90%50%10%1%

What each source reports

CVE List Score used 6.8

llama.cpp before b5649 Heap Out-of-Bounds Read via llama_vocab::impl::print_info

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.

Record
CVE-2025-71439
CVSS 4.0
6.8 MediumCVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
CVSS 3.1
5.5 MediumCVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
Weakness
CWE-125
Updated
2026-11-02
EUVDEUVD-2025-211201 6.8

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.

Record
EUVD-2025-211201
CVSS 4.0
6.8 MediumCVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
Updated
2026-11-02
GitHub AdvisoriesGHSA-4qpr-vwfh-qwp4 6.8

llama.cpp before b5649 contains a heap out-of-bounds read vulnerability in llama_vocab::impl:...

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.

Record
GHSA-4qpr-vwfh-qwp4
CVSS 4.0
6.8 MediumCVSS: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
CVSS 3.1
5.5 MediumCVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
Weakness
CWE-125
Updated
2026-11-02
NVD 6.8

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.

Record
CVE-2025-71439
CVSS 4.0
6.8 MediumCVSS: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
CVSS 3.1
5.5 MediumCVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
Weakness
CWE-125
Updated
2026-11-02
OSV 6.8

llama.cpp before b5649 Heap Out-of-Bounds Read via llama_vocab::impl::print_info

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.

Record
CVE-2025-71439
CVSS 4.0
6.8 MediumCVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
Weakness
CWE-125
Updated
2026-11-02
OSVUBUNTU-CVE-2025-71439 6.8

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.

Record
UBUNTU-CVE-2025-71439
CVSS 4.0
6.8 MediumCVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
CVSS 3.1
5.5 MediumCVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
Updated
2026-11-02

Affected products

Affected packages

Affected packages
PackageAffectedFixed
llama.cpp
Ubuntu:26.10
 All versions, plus listed versions
8681+dfsg-19721+dfsg-1ubuntu110438+dfsg-110438+dfsg-1ubuntu1
 None known
llama.cpp
Ubuntu:26.04:LTS
 All versions, plus listed versions
5882+dfsg-28064+dfsg-1ubuntu18681+dfsg-1
 None known

History

History
RecordedChangedSources
first indexedCVE List, EUVD, GitHub Advisories, NVD, OSV

Dates are when Every CVE recorded each change. Upstream dates may differ.

In short

CVE-2025-71439 is a medium vulnerability (CVSS 6.8) in Ggml-org Llama.cpp, published October 27, 2026. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.12%, the 1st percentile. No source names a fixed version.

References

Show 8 moreShow less
Pages All vulnerabilities Browse Docs Sources Guide Get API key
MoveEnter Open