CVE-2026-59707 LocalAI - Server-Side Request Forgery via POST /models/apply
Unauthenticated attackers can force LocalAI to fetch internal URLs via the /models/apply endpoint, leaking partial response data.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2026-59707?
CVE-2026-59707 is a critical server-side request forgery vulnerability in LocalAI, published on 7 Jul 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Official description
LocalAI contains an unauthenticated server-side request forgery vulnerability in the POST /models/apply endpoint that allows attackers to fetch arbitrary internal URLs. The endpoint passes unsanitized gallery URL fields directly to gallery.GetGalleryConfigFromURLWithContext without proper validation, enabling attackers to force the server to issue HTTP GET requests to private and loopback ranges with partial response content leaked through error messages.
Weakness: CWE-918
Which versions are affected by CVE-2026-59707?
3 open source packages ship an affected version, including github.com/mudler/LocalAI.
| Package | Affected | Fixed in |
|---|---|---|
github.com/mudler/LocalAI Go | All versions | 1.40.1-0.20260703213242-2cbb3c96b34d |
github.com/mudler/LocalAI Go | All versions | 1.40.1-0.20260703213242-2cbb3c96b34d |
github.com/mudler/LocalAI Go | All versions | 1.40.1-0.20260703213242-2cbb3c96b34d |
How do I fix CVE-2026-59707?
Upgrade github.com/mudler/LocalAI to 1.40.1-0.20260703213242-2cbb3c96b34d or later.
Upgrade, and send outbound requests through an allow-list that blocks internal addresses. More about Server-side request forgery.
Is CVE-2026-59707 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-59707 exists.
Public exploit code: 1 Nuclei template.
Sources
Also known as GHSA-8c5q-hx4g-qq23, GO-2026-6645.