LMDeploy is a toolkit for compressing, deploying, and serving LLMs. Prior to version 0.11.1, an insecure deserialization vulnerability exists in lmdeploy where torch.load() is called without the weights_only=True parameter when loading model checkpoint files. This allows an attacker to execute arbitrary code on the victim's machine when they load a malicious .bin or .pt model file. This issue has been patched in version 0.11.1.
References
Configurations
History
No history.
Information
Published : 2025-12-26 22:15
Updated : 2025-12-31 21:31
NVD link : CVE-2025-67729
Mitre link : CVE-2025-67729
CVE.ORG link : CVE-2025-67729
JSON object : View
Products Affected
internlm
- lmdeploy
CWE
CWE-502
Deserialization of Untrusted Data
