{"id":"CVE-2025-59953","published":"2026-09-16T16:17:03.010","lastModified":"2026-09-18T18:17:04.180","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.","cvssScore":9.8,"cvssSeverity":"CRITICAL","cvssVector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H","cwes":["CWE-502"],"vendors":[],"products":[],"references":[{"url":"https://github.com/InternLM/lmdeploy/releases/tag/v0.10.2","tags":[]},{"url":"https://github.com/InternLM/lmdeploy/security/advisories/GHSA-5h8j-6crg-7rmw","tags":[]},{"url":"https://github.com/InternLM/lmdeploy/security/advisories/GHSA-5h8j-6crg-7rmw","tags":[]}],"exploitRefs":[{"url":"https://github.com/InternLM/lmdeploy/releases/tag/v0.10.2","tags":[]},{"url":"https://github.com/InternLM/lmdeploy/security/advisories/GHSA-5h8j-6crg-7rmw","tags":[]},{"url":"https://github.com/InternLM/lmdeploy/security/advisories/GHSA-5h8j-6crg-7rmw","tags":[]}],"hasPoc":true,"ai":{"summary":"The vulnerability allows remote code execution due to improper deserialization of messages in the rpc server.","exploitability":"Exploitation is relatively easy as it requires sending maliciously crafted pickled data to the RPC server.","blast_radius":"If exploited, this could lead to full compromise of systems hosting the LMDeploy service.","remediation":"Update to version 0.10.2 or later which includes a patch for this vulnerability.","tags":["rce","deserialization","rpc","patch"],"model":"qwen2.5:7b-instruct","analyzedAt":"2026-09-23T08:46:07.812Z"}}