Skip to content
CVE Tracker
  • Newest CVEsPage
  • Recently added to KEVPage
  • Rising exploitation (EPSS)Page
  • Weekly digestPage
  • StatisticsPage
  • WatchlistPage
  • API documentationPage
  • MCP serverPage
  • Data sourcesPage
  • ChangelogPage
  • AboutPage
  • How it's builtPage
Back

CVE-2026-33625

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

High8.8EPSS 0.44%
Published
Updated
Reserved
Assigned by
GitHub_M

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.

Scores

CVSS scores of CVE-2026-33625, by who scored it
Scored byCVSSScoreVector
GitHub_MCNA (assigner)3.1High8.8CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
NVDNVD · Deferred3.1High8.8CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
EPSS (exploitation in 30 days)
0.44% · top 61% of all CVEs (6 Nov 2026)
CISA SSVC
Exploitation: poc · Automatable: no · Technical impact: total
Loading the EPSS history…

The list uses the assigner's score when there is one, then CISA's, then NVD's: High 8.8.

Weakness

Affected products

Affected packages

From reviewed advisories of the GitHub Advisory Database.

References (2)

Sources

The same CVE in the primary databases and aggregators, to check against.

Badge for a README or a ticket

CVE-2026-33625 badge
[![CVE-2026-33625](https://trackcve.app/badge/CVE-2026-33625.svg)](https://trackcve.app/cve/CVE-2026-33625)

Record last updated .