CyberSecBulletin
CVE-2025-53630NVD status: Deferred

llama.cpp: Heap-based Buffer Overflow

CVE-2025-53630 is a high-severity heap-based buffer overflow vulnerability in llama.cpp from Ggml-org, with a CVSS 4.0 base score of 8.9 out of 10. It can be exploited remotely over the network by an unauthenticated attacker, without any action from a user. Affected llama.cpp versions: < 26a48ad699d50b6268900062661bd22f3e792579. The record does not name a fixed version, so check the vendor advisory. There is no confirmed exploitation so far. EPSS estimates a 0.36% chance of exploitation within 30 days, which is low compared with other CVEs.

Next patch cycle

High severity with no evidence of exploitation so far.

What to do

  1. Check which version of llama.cpp you run against the affected versions below, and install the vendor's latest release.
  2. Read the advisory on github.com for workarounds and exact fixed builds.
  3. Until you can update, restrict network access to the affected service so it is not reachable from untrusted networks.

Affected products and versions

VendorProductAffected versionsFixed in
Ggml-orgllama.cpp< 26a48ad699d50b6268900062661bd22f3e792579Not stated

Version data comes from the organisation that assigned the CVE. Confirm against the vendor advisory before deciding you are unaffected.

How the severity score breaks down

Attack vector
NetworkCan be exploited remotely over a network, including the internet.
Attack complexity
LowNo special conditions are needed; the attack works reliably.
Attack requirements
NoneThe attack does not depend on a particular deployment or configuration.
Privileges required
NoneNo account or login is needed.
User interaction
NoneNo action by a victim is required.
Confidentiality (vulnerable system)
HighSerious loss of confidentiality on the vulnerable system.
Integrity (vulnerable system)
HighSerious loss of integrity on the vulnerable system.
Availability (vulnerable system)
HighSerious loss of availability on the vulnerable system.
Confidentiality (other systems)
NoneNo loss of confidentiality on systems beyond the vulnerable one.
Integrity (other systems)
NoneNo loss of integrity on systems beyond the vulnerable one.
Availability (other systems)
NoneNo loss of availability on systems beyond the vulnerable one.

Vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:P/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. Scored by CNA (github.com).

Official description

llama.cpp is an inference of several LLM models in C/C++. Integer Overflow in the gguf_init_from_file_impl function in ggml/src/gguf.cpp can lead to Heap Out-of-Bounds Read/Write. This vulnerability is fixed in commit 26a48ad699d50b6268900062661bd22f3e792579.

Source: the CVE record as published by its numbering authority, via NVD and CVE.org.

The underlying weakness: Heap-based Buffer Overflow

The product writes past the end of a buffer that was allocated on the heap.

How developers prevent this class of flaw

  • Pre-design: Use a language or compiler that performs automatic bounds checking.
  • Use an abstraction library to abstract away risky APIs. Not a complete solution.
  • Use automatic buffer overflow detection mechanisms that are offered by certain compilers or compiler extensions. Examples include: the Microsoft Visual Studio /GS flag, Fedora/Red Hat FORTIFY_SOURCE GCC flag, StackGuard, and ProPolice, which provide various mechanisms including canary-based detection and range/index checking. D3-SFCV (Stack Frame Canary Validation) from D3FEND discusses canary-based detection in detail.

More Heap-based Buffer Overflow vulnerabilities (CWE-122)

Timeline

  1. CVE record published.
  2. CISA published its SSVC assessment (exploitation: poc, automatable: no, technical impact: partial).
  3. Record last changed materially (score, affected products, status or references).

References

Other vulnerabilities in llama.cpp

All llama.cpp vulnerabilities

Recent Heap-based Buffer Overflow vulnerabilities

Spotted an error? Tell us.