Vulnerability · Published Sep 18, 2026 · Assigned by GitHub_M

CVE-2026-33625LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading

High 8.8

In plain English

A high-severity uncontrolled resource consumption in lmdeploy from InternLM. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.4% (higher than 39% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.

Severity (CVSS 3.1)8.8High

Exploit odds (30 days)0.4%top 61.4% of all CVEs

Exploited in the wildNo reportsCISA SSVC: poc

Attack toolingNoneno public tooling known

Official description

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions 012.1 through 0.12.2 contain a code injection vulnerability in `lmdeploy/pytorch/config.py` line 620 that allows an attacker to execute arbitrary Python code by publishing a malicious HuggingFace model with a crafted `quantization_config.quant_dtype` value. When a user loads the model with lmdeploy, the `quant_dtype` is passed to `eval(f'torch.{quant_dtype}')` without any validation. Version 0.12.3 contains a patch.

Affected packages & fixed versions

Ecosystem	Package	Fixed in	Advisory
PyPI	lmdeploy	0.12.3	GHSA-3hmm-rh5q-gwwr
PyPI	lmdeploy	0.12.3	PYSEC-2026-4078

Affected products (as reported)

Vendor	Product	Versions
InternLM	lmdeploy	>= 0.12.1, < 0.12.3

Weakness: CWE-400 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

References

https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr x_refsource_CONFIRM
https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3 x_refsource_MISC