AI and data stack advisories

Severe, 6 weeks1498Projects326

1498 severe, 6 weeks · 326 projects

LMDeployGHSA-7vc5-mjwp-c8fq

LMDeploy Improper Input Validation Vulnerability

LMDeploy

CVE-2025-3162 · Published Apr 3, 2025 · updated Jul 7, 2026

Medium5.3
No fix yet
GitHub advisory

A vulnerability was found in InternLM LMDeploy up to 0.7.1. It has been classified as critical. Affected is the function load_weight_ckpt of the file lmdeploy/lmdeploy/vl/model/utils.py of the component PT File Handler. The manipulation leads to deserialization. Attacking locally is a requirement. The exploit has been disclosed to the public and may be used.

Affected versions

PackageAffectedFixed in
lmdeploy
PyPI
<= 0.7.1No fix yet
Details and references
CVSS 3.1
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L
Severity from
GitHub (reviewed advisory)
Weakness
CWE-20, CWE-502
Also known as
CVE-2025-3162, PYSEC-2026-1577

More LMDeploy advisories

All LMDeploy
Critical 9.8Sep 16LMDeploy: remote code executionCVE-2025-59953fixed in 0.10.2High 7.8May 21lmdeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-outCVE-2026-46517fixed in 0.13.0High 7.8May 21LMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initializationCVE-2026-46432fixed in 0.13.0High 7.5Apr 21LMDeploy has Server-Side Request Forgery (SSRF) via Vision-Language Image LoadingCVE-2026-33626no fix yetHigh 8.8Dec 26, 2025lmdeploy vulnerable to Arbitrary Code Execution via Insecure Deserialization in torch.load()CVE-2025-67729fixed in 0.11.1Medium 5.3Apr 3, 2025InternLM LMDeploy code injection vulnerabilityCVE-2025-3163no fix yet