---
id: CVE-2025-66455
title: >-
  LMDeploy is a toolkit for compressing, deploying, and serving large language
  models
summary: >-
  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 m…
severity: critical
cvss: 9.8
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H'
cwe:
  - CWE-502
vendor: InternLM
product: lmdeploy
affected:
  - 'lmdeploy >= 0.9.2, < 0.16.0'
patched:
  - lmdeploy 0.16.0
published: '2026-09-18'
updated: '2026-09-18'
sourceUpdated: '2026-09-18T20:17:00.043'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2025-66455'
references:
  - url: >-
      https://github.com/InternLM/lmdeploy/commit/f05b4ad8bf2e2d84101a1d63b3c44fadd99223b2
    label: security-advisories@github.com
  - url: 'https://github.com/InternLM/lmdeploy/releases/tag/v0.16.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/InternLM/lmdeploy/security/advisories/GHSA-2vh9-42vm-xmv2
    label: security-advisories@github.com
  - url: 'https://github.com/advisories/GHSA-2vh9-42vm-xmv2'
  - url: 'https://github.com/InternLM/lmdeploy'
tags:
  - nvd
  - cve.org
  - ghsa
  - pip
  - osv
ssvc:
  exploitation: none
  automatable: 'yes'
  technicalImpact: total
  timestamp: '2026-09-18T19:54:08.825899Z'
aliases:
  - GHSA-2vh9-42vm-xmv2
ecosystem: pip
ingestedAt: '2026-09-18T17:46:41.532Z'
epss: 0.00702
epssPercentile: 0.51804
---

## Overview

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.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2025-66455)

Affected packages:

- `lmdeploy >= 0.9.2, < 0.16.0`

Patched in:

- `lmdeploy 0.16.0`

Source: https://github.com/advisories/GHSA-2vh9-42vm-xmv2
