Skip to main content
Public proof of conceptFix available

CVE-2026-33625

A public proof of concept exists; no exploitation attested.

  • Published2026-09-18
  • Fixed in0.12.3
  • WeaknessCWE-400

Am I affected?

With a domain, we read the version its public headers and pages expose. A distribution build can carry the fix under an older number.

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.

Timeline

dates in UTC
  1. 2026-09-1818:17CVE publishedNVD

Versions per branch

OSV
lmdeploy
BranchAffectedFixed in
0.12≥ 0.12.1, < 0.12.30.12.3

as publishedRanges as NVD and OSV publish them; a distribution’s build can be fixed under an older number.

Products declared by the CNA1 product

As GitHub_M, the CNA that assigned this CVE, declared them in the CVE record; versions as written.

  • InternLM lmdeploy
    Affected
    • = >= 0.12.1, < 0.12.3no fix stated

Source: the CNA record (CVE Program), © The MITRE Corporation,CVE Terms of UseCVE recordupdated 2026-09-18

Fix

  1. Update lmdeploy to 0.12.3

Detection

Nuclei template
none so far
ProjectDiscovery
Exploit-DB
no entry
Exploit-DB
Proof of concept
Metasploit modules
none listed
Metasploit Framework

Sources

  • NVDCVE recordread 2026-10-23
  • CISA-ADPSSVC decisionpoc2026-09-22
  • CNACVSS 3.18.8known since 2026-10-23
  • CNAweakness (CWE)CWE-400known since 2026-10-23
  • FIRST EPSSEPSS score0.4 %2026-11-18
  • OSVaffected packages—
  • GitHub AdvisoriesadvisoryHIGH2026-09-18

Identifiers

Technical details

CVSS v3 Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Published
2026-09-18
Last Modified
2026-09-24
NVD status
Deferred
GitHub Advisory
GHSA-3hmm-rh5q-gwwr

Frequently asked questions

What is CVE-2026-33625?

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.

Is CVE-2026-33625 actively exploited?

A proof-of-concept exploit exists for CVE-2026-33625, but active exploitation has not been confirmed at this time.

What is the CVSS score for CVE-2026-33625?

CVE-2026-33625 has a CVSS v3 base score of 8.8 (HIGH severity), with vector string CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H.

Running lmdeploy? Hear about its next vulnerability

Surveillance reads the versions in your stack — a domain, a lockfile or a list — checks each CVE against the ranges its sources publish, and alerts you when one you run is exploited or enters CISA KEV. A version it cannot read stays « version not verified ».

No credit card required · No monthly cap on website lookups · 250 API requests/month

Sources for this record