Skip to content

CVE-2025-59953 LMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeploy

CriticalCVSS 9.8 EPSS 0.88%
Published
Updated
Exploit status
No public exploit known
CISA SSVC
Proof of concept public · automatable: yes · technical impact: total

Description

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.1 and prior to version 0.10.2, the LMdeploy implements an rpc server (AsyncRPCServer in zmq_rpc.py) for supporting the RPC communications. In its core functionality call_and_response(), I found it will directly use the pickles.loads() to deserialize the received messages without any sanitization, hence resulting in a remote code execution vulnerability by this RPC server. Version 0.10.2 contains a patch.

Upgrade path

2 packages: 1 fixed.

PackageAffectedFixed inStatus
pypi lmdeploy(PyPI) = 0.9.1; = 0.9.2; = 0.9.2.post1; = 0.10.0 +2 more 0.10.2 Fixed
InternLM lmdeploy = >= 0.9.1, < 0.10.2 No fix listed Versions listed

Affected products

ProductAffected versionsFixed in
pypi lmdeploy(PyPI) >= 0.9.1, < 0.10.2 0.10.2
pypi lmdeploy(PyPI) = 0.10.0 No fix listed
pypi lmdeploy(PyPI) = 0.10.1 No fix listed
pypi lmdeploy(PyPI) = 0.9.1 No fix listed
pypi lmdeploy(PyPI) = 0.9.2 No fix listed
pypi lmdeploy(PyPI) = 0.9.2.post1 No fix listed
InternLM lmdeploy = >= 0.9.1, < 0.10.2 No fix listed

References