- CVSS
- HIGH · 7.8v3.1CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
- Published
- 2026-06-10
- Weakness
- CWE-94, CWE-915, CWE-1188
- Source
- nvd.nist.gov/vuln/detail/CVE-2026-46517
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.
References
- https://github.com/InternLM/lmdeploy/blob/v0.13.0/lmdeploy/version.py
- https://github.com/InternLM/lmdeploy/commit/81be52961aa324fd5cd3cacebffba1ba051bc107
- https://github.com/InternLM/lmdeploy/releases/tag/v0.13.0
- https://github.com/InternLM/lmdeploy/security/advisories/GHSA-9xq9-36w5-q796
- https://github.com/github/advisory-database/pull/4511
- https://github.com/pypa/advisory-database/tree/main/vulns/lmdeploy/PYSEC-2026-2608.yaml
How GTK Cyber trains on this
AI security training at GTK Cyber covers the LLM and ML-pipeline vulnerability classes that vulnerabilities like CVE-2026-46517 fall into. Our hands-on courses are taught by Charles Givre and other practitioners who break and defend production AI systems.