CVE-2025-66455: Insecure Deserialization (CVSS 9.8)

Description

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.

Is your site exposed to CVE-2025-66455?

Run a free security scan — no signup, results in seconds.

CVSS v3.1 Score

9.8

CRITICAL

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

EPSS — Exploit Prediction

0.0069

Probability of exploitation

0.52%

Percentile rank

EPSS estimates the probability that this vulnerability will be exploited in the wild within the next 30 days. A higher score means more likely to be exploited.

Weakness Type (CWE)

CWE-502 Deserialization of Untrusted Data

References

Other References

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

Frequently Asked Questions

What is 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. It has a CVSS v3.1 base score of 9.8 (CRITICAL).

How severe is CVE-2025-66455? +

CVE-2025-66455 has a CVSS v3.1 score of 9.8 out of 10, rated CRITICAL. This is a critical vulnerability that should be patched immediately. The EPSS score is 0.0069, placing it in the 1th percentile for exploitation probability.

How do I check if I'm vulnerable to CVE-2025-66455? +

You can use Secably's free Website Scanner to check your website for known vulnerabilities. For infrastructure scanning, use the Port Scanner to identify exposed services that may be affected. Check the vendor advisories linked above for specific patch and version information.

Related Vulnerabilities

CVE-2026-10748

An authenticated user with the nx-licensing-create privilege can upload a specially crafted license file to execute arbitrary operating system commands …

CVE-2026-101169

In affected versions of Octopus Server, an authenticated user with permissions to edit an Environment or Project can set specifically …

CVE-2026-44126

SEPPmail Secure Email Gateway before version 15.0.4 insecurely deserializes untrusted data, which can be reached from the new GINA UI …

CVE-2026-11857

Quanos SCHEMA ST4 on-premises contains a local privilege escalation vulnerability in the Client Update Service due to insecure deserialization in …

CVE-2026-47161

RELATE is a web-based courseware package. Prior to commit d66ba5659b459bf1ba56b7109b5f9ecf197cbefb, RELATE LMS configures its Celery workers to accept and deserialize …

CVE-2025-1077

A security vulnerability has been identified in the IBL Software Engineering Visual Weather and derived products (NAMIS, Aero Weather, Satellite …

CVE-2025-66455