CVE-2026-76841
HIGH
CVSS 8.8
CVE-2026-76841 HIGH
Qu'est-ce que c'est ?
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 trust_remote_code=True as a literal or as an unconditional default: RerankModel._get_tokenizer in xinference/model/rerank/core.py, SentenceTransformerRerankModel.load in xinference/model/rerank/sentence_transformers/core.py, SentenceTransformerEmbeddingModel.load in xinference/model/embedding/sentence_transformers/core.py, FlagEmbeddingModel.load in xinference/model/embedding/flag/core.py, and two sites in xinference/model/llm/transformers/core.py where PytorchModel._sanitize_model_config and PytorchModel._get_components default the value to True. Because a caller with model launch access can register a model whose type is unknown and supply an arbitrary model path, the server reaches _auto_detect_type and then AutoTokenizer.from_pretrained, which imports and executes Python declared by the model directory's own tokenizer_config.json auto_map, running attacker-supplied code with the privileges of the worker process. Version 2.12.0 gates every site behind allow_trust_remote_code and the XINFERENCE_TRUST_REMOTE_CODE setting, permitting remote code only for bundled built-in models.
Technologies affectées
Non précisé par la fiche NVD.
Recommandation
Vérifie la disponibilité d'un correctif pour la technologie concernée et applique-le dès que possible. Consulte la fiche NVD pour les références officielles du fournisseur.
Exploits publics connus (0)
Aucun proof-of-concept public trouvé sur GitHub à ce jour. Ça peut changer : les PoC sont souvent publiés plusieurs jours ou semaines après la divulgation.