LightLLM through 1.2.0 visual_only deployments expose an unauthenticated RPyC service with allow_pickle enabled that deserializes attacker-supplied arguments in the remote_infer_images method. Attackers can reach the visual RPyC port and pass objects with __reduce__ methods to execute arbitrary code with service account privileges.
References
Configurations
No configuration.
History
30 Sep 2026, 20:17
| Type | Values Removed | Values Added |
|---|---|---|
| References | () https://github.com/ModelTC/LightLLM/issues/1610 - |
30 Sep 2026, 15:22
| Type | Values Removed | Values Added |
|---|---|---|
| New CVE |
Information
Published : 2026-09-30 15:22
Updated : 2026-09-30 20:17
NVD link : CVE-2026-103395
Mitre link : CVE-2026-103395
CVE.ORG link : CVE-2026-103395
JSON object : View
Products Affected
No product.
CWE
CWE-502
Deserialization of Untrusted Data
