# CVE-2026-59707: LocalAI - SSRF

> Live exploitation tracking for CVE-2026-59707 by the CrowdSec Live Exploit Tracker, built from real-world attack data crowdsourced from CrowdSec's global network.

Page: https://tracker.crowdsec.net/cves/CVE-2026-59707

## Key facts

- **Exploitation phase:** Limited Exploitation. The vulnerability is known but shows very limited attacker interest or exploitation activity.
- **CVSS score:** 9.2
- **Public exploit available:** Yes
- **Affected products:** LocalAI
- **Weaknesses:** CWE-918 (Server-Side Request Forgery (SSRF))
- **Tags:** ai_mcp, web_application

## 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.

## CrowdSec analysis

[CVE-2026-59707](https://euvd.enisa.europa.eu/vulnerability/CVE-2026-59707) is a critical unauthenticated server-side request forgery (SSRF) vulnerability in LocalAI’s `POST /models/apply` endpoint. Attackers can make the server send requests to arbitrary internal, private, or loopback URLs, potentially exposing partial response data through error messages and enabling reconnaissance of otherwise inaccessible systems.

CrowdSec has been tracking this vulnerability and its exploits since 26th of October 2026.

CrowdSec network observations suggest that most exploitation of CVE-2026-59707 involves focused reconnaissance to identify viable targets. Attackers typically tailor their campaigns based on system exposure and configuration. It is unlikely that a given attack is accidental.
Data from the CrowdSec community also indicates a gradual decrease in attacks targeting CVE-2026-59707. While still present in the wild, exploitation levels have dropped noticeably week-over-week. This may signal that the vulnerability is becoming less relevant or that defenses are improving fast enough for attackers to lose interest.

Attackers target the `POST /models/apply` endpoint and supply a crafted URL in the JSON `url` field, causing the server to make requests to attacker-controlled or internal network resources. The fetched result may be reflected in the model-apply job status.

## Timeline

- 2026-07-07: CVE Published. CVE-2026-59707 is published to NVD.
- 2026-10-26: Rule Released. CrowdSec releases a rule to detect CVE-2026-59707 exploitation attempts against the CrowdSec Network.
- 2026-10-28: CrowdSec First Seen. CrowdSec observes in-the-wild exploitation of CVE-2026-59707 for the first time.

## References

- https://github.com/mudler/LocalAI/issues/10665
- https://github.com/mudler/LocalAI
- https://github.com/mudler/LocalAI/commit/f9b968e19d7cbc556d59dceb2e0e450b828a3fda
- https://www.vulncheck.com/advisories/localai-server-side-request-forgery-via-post-models-apply
- https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2026/CVE-2026-59707.yaml

## Live exploitation data

The [Live Exploit Tracker page](https://tracker.crowdsec.net/cves/CVE-2026-59707) follows observed exploitation activity for CVE-2026-59707. [Request an API key](https://tracker.crowdsec.net/request-api-key) for full access to its exploitation timeline, attacking IP addresses and blocklists.