CVE-2024-11449 Server-Side Request Forgery in haotian-liu/llava
An attacker can exploit insufficient validation in the predict endpoint to access internal networks or cloud metadata, exposing sensitive data.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- None listed
What is CVE-2024-11449?
CVE-2024-11449 is a high-severity server-side request forgery vulnerability in haotian-liu/llava, published on 20 Mar 2025. An attacker can exploit it over the internet without logging in, with no action from the victim.
Official description
A vulnerability in haotian-liu/llava version 1.2.0 (LLaVA-1.6) allows for Server-Side Request Forgery (SSRF) through the /run/predict endpoint. An attacker can gain unauthorized access to internal networks or the AWS metadata endpoint by sending crafted requests that exploit insufficient validation of the path parameter. This flaw can lead to unauthorized network access, sensitive data exposure, and further exploitation within the network.
Weakness: CWE-918
Which versions are affected by CVE-2024-11449?
It affects 1 product.
Products include Hliu Large Language and Vision Assistant.
How do I fix CVE-2024-11449?
No fix has been published for CVE-2024-11449 yet. Limit exposure of the affected software until one is available.
Upgrade, and send outbound requests through an allow-list that blocks internal addresses. More about Server-side request forgery.
Is CVE-2024-11449 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2024-11449 exists.
Other vulnerabilities in haotian-liu/llava
Sources
Also known as GHSA-22h9-9rvv-q9qg.