CVE-2026-33625: fallo de gravedad alta en InternLM lmdeploy
LMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loading
Publicada el · Actualizada el
Aviso por correo cuando entre en el KEV, tenga PoC pública o suba de nota.
21Vexday Risk Score
Sin señal de explotación. Ningún artefacto público de explotación conocido hasta ahora.
ssvc Trackcvss 8.8epss 0.4%
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:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Productos afectados
InternLM · lmdeployCVEs relacionadas — InternLM lmdeploy
En el mismo producto, de las más peligrosas a las menos.
CVE-2026-33626HIGHLMDeploy Vulnerable to Server-Side Request Forgery (SSRF) via Vision-Language Image LoadingEPSS 45.3%CVE-2026-76850CRITICALLMDeploy Remote Code Execution via Unsafe Pickle Deserialization in the Disaggregated Serving Peer ConnectorEPSS 1.3%CVE-2025-66455CRITICALLMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.pyEPSS 1.3%CVE-2026-119706CRITICALLMDeploy 0.9.0 through 0.18.0 Unauthenticated RCE via /update_weights Pickle DeserializationEPSS 1.0%CVE-2025-59953CRITICALLMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeployEPSS 0.9%CVE-2026-92971HIGHInternLM LMDeploy through 0.17.0 Assertion Denial of ServiceEPSS 0.7%