CVE-2026-108836: Ollama server-side request forgery in Pull API

CVE-2026-108836 · Severity: high · CVSS 7.3 · Published 2026-10-12

Ollama is a tool for running large language models locally. A flaw in the Pull API's model name validation allows remote attackers to perform server-side request forgery (SSRF), potentially enabling them to access internal services or resources that the Ollama server can reach. The vulnerability is remotely exploitable and has been publicly disclosed.

Technologies: Ollama. Vendors: Ollama.

Quick answers

Is CVE-2026-108836 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-108836 being exploited in the wild.
How severe is CVE-2026-108836?
High severity, with a CVSS score of 7.3, published 12 October 2026.
Which products and versions are affected?
Ollama up to 0.34.1.
Is there a fix for CVE-2026-108836?
Yes: fixed in 0.34.2. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-108836 attacked?
It is attacked over the network.
Am I affected by CVE-2026-108836?
Check whether you run Ollama at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

  • Fixed in: 0.34.2

Technical details

The vulnerability exists in the IsValid function of types/model/name.go in the Pull API component. A flaw in model name validation leads to server-side request forgery. The attack is remotely exploitable with no authentication or user interaction required. An attacker can manipulate the vulnerability to cause the server to make unintended HTTP requests to internal systems.

Affected products

  • Ollama Ollama up to 0.34.1

Timeline

  • 2026-10-12: disclosed
  • 2026-10-12: patched: Patch dfabde4539e42ba1e1eab50a3a50b88aea7958a0

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.