CVE-2026-46517

LMDeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out

Published Jun 9, 2026 CVSS 3.1 · 7.8 High EPSS < 1% KEV · Not listed
Description
LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trust_remote_code=True" enables HF supply-chain RCE without user opt-in. Version 0.13.0 patches the issue.

Analysis

AI-generated

Technical impact

LMDeploy, a toolkit for compressing, deploying, and serving large language models, has a hardcoded configuration that sets trust_remote_code=True in versions 0.12.3 and earlier. This setting allows automatic loading of code from Hugging Face model repositories without user opt‑in, creating a potential remote code execution path. Attackers who can supply or modify a model payload can have malicious code executed during model loading, with the same privileges as the LMDeploy process.

Affected Systems

The affected product is InternLM’s LMDeploy, specifically all releases up to and including 0.12.3. The issue is fixed in version 0.13.0, so upgrading removes the risk.

Risk and Exploitability

The CVSS score of 7.8 indicates a high‑severity flaw. The EPSS score is 0.00148, indicating a very low probability of exploitation, and the issue is not listed in the CISA KEV catalog, but the nature of the flaw—automatic loading of untrusted code—creates a high likelihood of exploitation by attackers with access to the model repository or deployment environment. The attack vector is remote, leveraging the model loading mechanism; an attacker can supply or modify a model definition that contains malicious code which will be executed during deployment.

Remediation

No solution or workaround provided in the CVE record.

OpenCVE Recommended Actions

  • Configure LMDeploy to set trust_remote_code=False in all deployment configurations.
  • Manually review and verify any model code before execution within LMDeploy.
  • Monitor InternLM’s release channel and apply an updated version once a patch that removes the hardcoded flag becomes available.

Generated by OpenCVE AI on August 31, 2026 at 15:59 UTC.

Tracking

Sign in to view the affected projects.

Advisories

Source ID Title
Github GHSA Github GHSA GHSA-9xq9-36w5-q796 lmdeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out
History
Type Values Removed Values Added
Description LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trust_remote_code=True" enables HF supply-chain RCE without user opt-in. At time of publication, there are no publicly available patches. LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trust_remote_code=True" enables HF supply-chain RCE without user opt-in. Version 0.13.0 patches the issue.
References
Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'total'}, 'version': '2.0.3'}

ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'none', 'Technical Impact': 'total'}, 'version': '2.0.3'}

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'total'}, 'version': '2.0.3'}

Type Values Removed Values Added
First Time appeared Internlm
Internlm lmdeploy
Vendors & Products Internlm
Internlm lmdeploy
Type Values Removed Values Added
Description LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, hardcoded "trust_remote_code=True" enables HF supply-chain RCE without user opt-in. At time of publication, there are no publicly available patches.
Title LMDeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out
Weaknesses CWE-1188
CWE-915
CWE-94
References
Metrics cvssV3_1

{'score': 7.8, 'vector': 'CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H'}

Weaknesses
  • CWE-1188

    Initialization of a Resource with an Insecure Default

  • CWE-915

    Improperly Controlled Modification of Dynamically-Determined Object Attributes

  • CWE-94

    Improper Control of Generation of Code ('Code Injection')

Sources

Status
PUBLISHED
Assigner
GitHub_M
Published
Updated
Aug 31, 2026, 2:34 PM
Reserved
May 14, 2026, 7:12 PM
Status
Deferred
Published
Jun 10, 2026, 12:16 AM
Modified
Aug 31, 2026, 3:17 PM