CVE-2025-66455

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.2 and prior to version 0.16.0, LMDeploy's PyTorch DistServe/PD-disaggregation control plane used `recv_pyobj()` to deserialize messages received through a ZeroMQ PULL socket. PyZMQ implements `recv_pyobj()` using Python pickle deserialization, which can execute arbitrary code while reconstructing an object. The peer address used by the receiver was supplied through the `POST /distserve/p2p_connect` HTTP endpoint. An attacker who could reach an affected DistServe API server could cause the server to connect to an attacker-controlled ZeroMQ endpoint and deserialize a crafted pickle payload. API-key authentication is not enabled unless the operator explicitly configures it. As a result, affected DistServe deployments without API keys allowed unauthenticated remote code execution with the privileges of the LMDeploy serving process. This issue affects the PyTorch backend when PD-disaggregation/DistServe is enabled. Ordinary deployments that do not use the affected disaggregated-serving path do not expose this data flow. The fix was released in LMDeploy 0.16.0. Users who cannot upgrade immediately should prevent untrusted clients from reaching `/distserve/*` endpoints, restrict the DistServe HTTP and ZeroMQ control planes to trusted cluster networks, configure API-key authentication, and block arbitrary outbound ZeroMQ connections from serving nodes. These measures reduce exposure but do not make pickle deserialization safe.

• CWE-502

• https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2
• https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0
• https://github.com/InternLM/lmdeploy/security/advisories/GHSA-2vh9-42vm-xmv2

Published

Sep 18, 2026

Last Modified

Sep 23, 2026

Status

Awaiting Analysis

• How Strix found a critical auth bypass in etcdStrix autonomously discovered a critical authentication bypass in etcd, later designated CVE-2026-33413.
• Autonomous PentestingAI agents that find and validate exploitable vulnerabilities like this one across your applications.
• PR ReviewsPentest every pull request so vulnerable code is caught before it ships to production.
• AI Penetration TestingHow AI-driven penetration testing continuously covers your attack surface.

• CVE-2025-6645PDF-XChange Editor U3D File Parsing Use-After-Free Remote Co…7.8
• CVE-2025-66450LibreChat is a ChatGPT clone with additional features. In ve…5.4
• CVE-2025-66451LibreChat is a ChatGPT clone with additional features. In ve…6.5
• CVE-2025-66452LibreChat is a ChatGPT clone with additional features. In ve…6.1
• CVE-2025-66453Rhino is an open-source implementation of JavaScript written…7.5
• CVE-2025-66454Arcade MCP allows you to to create, deploy, and share MCP Se…6.5
• CVE-2025-66456Elysia is a Typescript framework for request validation, typ…9.8
• CVE-2025-66457Elysia is a Typescript framework for request validation, typ…8.8
• CVE-2025-66458Lookyloo is a web interface that allows users to capture a w…6.1
• CVE-2025-66459Lookyloo is a web interface that allows users to capture a w…6.1
• CVE-2025-6646PDF-XChange Editor U3D File Parsing Use-After-Free Informati…3.3
• CVE-2025-66460Lookyloo is a web interface that allows users to capture a w…6.1