Skip to main content

CVE-2026-133498mozilla-ai llamafile out-of-bounds read

Low priorityMedium5.3EPSS not scored yet

CVE-2026-133498 is a out-of-bounds read in mozilla-ai llamafile, rated medium (5.3 on CVSS 4.0).

Why this priority: Medium severity, with no exploitation signal.

Published · Updated · NVD status: Deferred

What it is

Code reads past the end of a buffer, leaking memory contents or crashing. Code reads or writes outside the memory it was given, which can corrupt the process.

How it is reached

An attacker can exploit it over the network, without an account, but someone has to act first, such as opening a file or following a link.

What an attacker gains

The direct impact is limited: partial access to data or a partial outage. In practice, this kind of flaw lets an attacker crash the process, read memory, or in the worst case run code.

Is it being exploited?

There is no public report of exploitation yet.

What to do

  1. Upgrade mozilla-ai llamafile to 0.10.6 or later.
  2. Until it is patched, keep it off the internet or behind a VPN, allowlist or authenticating proxy.
  3. Warn the people who use it, and filter files and links on the way in.

Description from NVD

A flaw has been found in mozilla-ai llamafile up to .0.10.5. This issue affects the function llamafile_open_zip of the file llamafile/llamafile.c of the component Container Parser. This manipulation of the argument size causes out-of-bounds read. The attack is possible to be carried out remotely. Upgrading to version 0.10.6 is capable of addressing this issue. Patch name: 5ac1c4dbb76737fee71a54922df95342c667ba6f. It is advisable to upgrade the affected component.

What this means for your architecture.

A CVE names a product. Whether it matters depends on where that product sits in your systems: what reaches it, which trust boundary it guards and what it can reach once compromised. Veriom maps your components, their relationships and your trust boundaries from your code, cloud and delivery evidence, so you can see where a component like this sits. The radar itself does not look at your systems.

Components it touches

  • Native code: operating systems, browsers and firmware. The flaw sits wherever native code reads untrusted data.
  • Network appliances and media or file parsers. The flaw sits wherever native code reads untrusted data.

Trust boundaries in play

  • Internet-facing entry points. No account is needed. Any instance reachable from outside your perimeter, through a load balancer, gateway, VPN or exposed admin port, is in scope.
  • People who open files and links. Mail, chat and browsers are the delivery path.
  • Wherever native code reads untrusted data. An attacker who crosses it can crash the process, read memory, or in the worst case run code.

Questions to answer about your own systems

  • Where does mozilla-ai llamafile run in your systems, and at which version?
  • Which of those instances can be reached from the internet or partner networks?
  • What can it reach once compromised: which data stores, secrets and services?

14-day free trial

Understand how your systems connect.

See your services, dependencies and trust boundaries in one place, with evidence from your connected environment.

Sources

  • CVE records and CVSS scores from the NVD CVE API. This product uses the NVD API but is not endorsed or certified by the NVD.
  • Exploitation status from CISA’s Known Exploited Vulnerabilities catalog.
  • Exploit likelihood from FIRST’s Exploit Prediction Scoring System (EPSS), refreshed daily.
  • Explanations are written by Veriom from fixed templates over each record’s data, not by a person reviewing each CVE. Descriptions are quoted from the records as published.