---
cve: "CVE-2026-33625"
severity: "HIGH"
cvss: 8.8
epss: "0.2%"
vendor: "InternLM"
kev: false
exploited: false
published: "2026-09-18 18:17:06"
tags: [cve, security, high]
source: tsecurity.de CVE-Dossier
exported: "2026-09-22T09:56:53+02:00"
---

# CVE-2026-33625

> 8.8 HIGH · 🧪 PoC

## Beschreibung

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.

## CVSS-Vektor

```
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
```

| Metrik | Wert | Bewertung |
|---|---|---|
| AV Angriffsvektor | Netzwerk | bad |
| AC Komplexität | Gering | bad |
| PR Privilegien | Keine | bad |
| UI Interaktion | Erforderlich | good |
| S Scope | Unverändert | good |
| C Vertraulichkeit | Hoch | bad |
| I Integrität | Hoch | bad |
| A Verfügbarkeit | Hoch | bad |

## Patch verfügbar (OSV)

- 8ea459f49ed9cd943481073011424919e31e3e3b (Commit)

## Referenzen

- <https://github.com/InternLM/lmdeploy/security/advisories/GHSA-3hmm-rh5q-gwwr>
- <https://github.com/InternLM/lmdeploy/releases/tag/v0.12.3>

---
_Exportiert aus dem [tsecurity.de CVE-Dossier](https://tsecurity.de/cve?cve=CVE-2026-33625) · Datenquellen: EUVD (ENISA), NVD, OSV, CISA KEV, FIRST EPSS, Exploit-DB, BSI BITS_