Vulnerability · Published Sep 30, 2026 · Assigned by VulnCheck
CVE-2026-103243LightLLM through 1.2.0 Server-Side Request Forgery via multimodal endpoints
Medium 5.8
In plain English
A medium-severity server-side request forgery in LightLLM from ModelTC. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.3% (higher than 24% of all known vulnerabilities).
Severity (CVSS 3.1)5.8Medium
Exploit odds (30 days)0.3%top 76.0% of all CVEs
Exploited in the wildNo reportsCISA SSVC: poc
Attack toolingNoneno public tooling known
Official description
LightLLM through 1.2.0 fails to validate image_url and audio_url parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error responses revealing internal network topology.
Affected products (as reported)
Weakness: CWE-918 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:N/A:N
References
- GitHub Issue #1608 issue-tracking
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/utils/multimodal_utils.py#L82-L91 technical-description
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/multimodal_params.py#L149 technical-description
- https://github.com/ModelTC/LightLLM product
- VulnCheck Advisory: LightLLM through 1.2.0 Server-Side Request Forgery via multimodal endpoints third-party-advisory