Sign in
Deutsch English Español Polski
Medium

InternLM LMDeploy code injection vulnerability

  • Disclosed: 03.04.2025
  • Updated: 11.10.2026 13:19

A vulnerability was found in InternLM LMDeploy up to 0.7.1. It has been declared as critical. Affected by this vulnerability is the function Open of the file lmdeploy/docs/en/conf.py. The manipulation leads to code injection. It is possible to launch the attack on the local host. The exploit has been disclosed to the public and may be used.

Affected software

Click a technology to see all vulnerabilities affecting it.

PlatformPackageVulnerable versionsFixed in
Python (PyPI) lmdeploy <= 0.7.1 no fix

References

CVSS score 5.3/10
CVSS vector
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L
Weakness (CWE)
CWE-74, CWE-94
Platform
Python (PyPI)
Package
lmdeploy
Disclosed
03.04.2025
Updated
11.10.2026 13:19

Sources and licenses

OSV / GitHub Advisory Database - PYSEC-2026-1579 - PyPA Advisory Database; CC BY 4.0

OSV / GitHub Advisory Database - GHSA-jfvg-qm4p-473x - CC BY 4.0

Vulnerability history: lmdeploy

All published vulnerabilities of this package (Python (PyPI)), newest first. Shows whether this is a one-off slip or a package that needs watching.

All (10) in the archive
Total entries
10
First → latest
04.2025 09.2026
Without a fix
3
Exploited in attacks
0

Severity breakdown

  • Critical 2
  • High 6
  • Medium 2

Disclosures per year

Severity:
  1. LMDeploy has an SSRF bypass
    High · 7.5 ghsa-39wr-7q6h-cf68 >= 0.12.3, < 0.15.0 0.15.0
  2. LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading
    High · 8.8 CVE-2026-33625 >= 0.12.1, < 0.12.3 0.12.3
  3. LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py
    Critical · 9.8 CVE-2025-66455 >= 0.9.2, < 0.16.0 0.16.0
  4. LMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeploy
    Critical · 9.8 CVE-2025-59953 >= 0.9.1, < 0.10.2 0.10.2
  5. lmdeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out
    High · 7.8 CVE-2026-46517 < 0.13.0 0.13.0
  6. LMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initialization
    High · 7.8 CVE-2026-46432 < 0.13.0 0.13.0
  7. LMDeploy has Server-Side Request Forgery (SSRF) via Vision-Language Image Loading
    High · 7.5 CVE-2026-33626 <= 0.12.2 no fix
  8. lmdeploy vulnerable to Arbitrary Code Execution via Insecure Deserialization in torch.load()
    High · 8.8 CVE-2025-67729 < 0.11.1 0.11.1
  9. InternLM LMDeploy code injection vulnerability this entry
    Medium · 5.3 CVE-2025-3163 <= 0.7.1 no fix

Know which vulnerabilities affect your application

With the DockRay integration and software inventory enabled, DockRay checks reported component versions against known vulnerabilities and notifies you about confirmed matches. Coverage depends on the supported technology and available advisory data.

Connect your application
DockRay newsletter

Application security in your inbox

Important vulnerabilities, actionable guidance and practical advice for your applications. Choose your technologies and follow what matters to your project.

  • A daily digest of new vulnerabilities, critical ones included.
  • Notifications about important updates.
  • Practical advice on application security and monitoring.

Leave empty to receive updates for every technology.

Privacy policy

Free to subscribe. Unsubscribe with one click in any email.