llama-factory-webui-rce-cve-2026-58116-trust-remote-code-model-path-injection

CVE-2026-58116 — RCE en WebUI de LLaMA-Factory mediante trustremotecode Prueba de concepto para CVE-2026-58116 CVSS 9.8 Crítico: ejecución remota de código en LLaMA-Factory = v0.9.5. La WebUI de LLaMA-Factory tiene hardcodeado trustremotecode=True al cargar un modelo...

CVE-2026-5817-PoC

CVE-2026-5817: RCE / Escape de contenedor a host en Docker Model Runner Cualquier contenedor en un host de Docker Desktop 4.40.0 a 4.67.x puede ejecutar código en el host con dos peticiones HTTP. Sin montaje de socket, sin --privileged, sin capabilities. Cómo funciona Cualquier contenedor puede...

CVE-2026-7669-PoC

CVE-2026-7669 O SGLang silenciosamente substitui trustremotecode=False por True e re-invoca AutoTokenizer.frompretrained sempre que o transformers v5 retorna um objeto TokenizersBackend. Um modelo com um tokenizerclass personalizado e um automap apontando para tokenizer.py alcança execução...

CVE-2026-22807_Range

🛡️ CVE-2026-22807 Simulazione: Range di Target RCE per Supply Chain AI ⚠️ Disclaimer / Avvertenza Questo progetto è destinato esclusivamente a scopi di ricerca e didattica sulla sicurezza , per dimostrare il principio degli attacchi alla supply chain durante il caricamento di modelli AI Remote Code...

FULL-ANALYSIS---CVE-2026-45829-ChromaDB-

CVE-2026-45829 - ChromaDB Pre-Auth RCE CVSS 10.0 ⚠️ AVVERTENZA Questo repository contiene una PoC FUNZIONALE per CVE-2026-45829. SOLO per test autorizzati in ambienti controllati. Descrizione Vulnerabilità critica in ChromaDB versioni 1.0.0 a 1.5.8 che consente a un attaccante non autenticato di...

CVE-2026-76841

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

EUVD-2026-64847

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

CVE-2026-76841

Xinference (by XorbitsAI) through version 2.11.0 unconditionally sets trust_remote_code=True at six model loader call sites spanning the rerank, embedding, and LLM modules. An attacker with model launch access can register a model with an unknown type and an arbitrary path, causing AutoTokenizer....

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

Xinference loads models with Hugging Face remote code execution unconditionally enabled, and before version 2.12.0 exposes no setting to disable it. Six loader call sites pass trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

Exploit for CVE-2026-47117

CVE-2026-47117: OpenMed Unauthenticated Remote Code Execution...

CVE-2026-45804

Diffusers is the a library for pretrained diffusion models. Prior to 0.38.0, Diffusers' DiffusionPipeline.frompretrained flow can bypass the trustremotecode guard because download validates modelindex.json and custom pipeline code before later loading from a cached folder that can change, allowin...

ChromaDB has a code injection vulnerability

A code injection vulnerability in version 0.4.17 or later of the ChromaDB Python project allows an authenticated attacker to run arbitrary code on the server by sending a malicious model repository and trustremotecode set to true in...

EUVD-2026-35873

LMDeploy is a toolkit for compressing, deploying, and serving large language models. In versions 0.12.3 and prior, LMDeploy is vulnerable to arbitrary code execution through hardcoded "trustremotecode=True" in multiple HuggingFace model-loading call sites. At time of publication, there are no...

CVE-2026-44513

A flaw was found in Diffusers, a library for pretrained diffusion models. A remote attacker could exploit a bypass in the trustremotecode mechanism within the DiffusionPipeline.frompretrained function. This vulnerability allows for arbitrary remote code execution, even when the user explicitly se...

CVE-2026-4944

vllm-project/vllm version 0.14.1 contains a vulnerability where the trustremotecode=True parameter is hardcoded in two model implementation files vllm/modelexecutor/models/nemotronvl.py and vllm/modelexecutor/models/kimik25.py. This bypasses the user's explicit --trust-remote-code=False setting,...

CVE-2026-45829

A pre-authentication, code injection vulnerability in version 1.0.0 or later of the ChromaDB Python project allows an unauthenticated attacker to run arbitrary code on the server by sending a malicious model repository and trustremotecode set to true in...

CVE-2026-44513 Diffusers: `trust_remote_code` bypass via `custom_pipeline` and local custom components

Diffusers is the a library for pretrained diffusion models. Prior to 0.38.0, a trustremotecode bypass in DiffusionPipeline.frompretrained allows arbitrary remote code execution despite the user passing trustremotecode=False or omitting it, which is the default. The vulnerability has three variant...

sglang 注入漏洞

SGLang is a programming language and runtime system developed by SGL-project, aimed at accelerating large model inference. Versions of SGLang 0.5.9 and earlier contained a injection vulnerability. This vulnerability stemmed from the operation of the gettokenizer function in the...