CVE-2026-76841High· 8.8▾ TwilightXinference 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 de…
▾ Twilight zone — High severity, or a signal on a lesser flaw
impact 48.4 · likelihood 0.2 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Stakeholder-Specific Vulnerability Categorization from CISA's ADP record at CVE.org: whether exploitation is observed, whether an attack can be automated, and how much of the system is at stake.
Exploit-prediction probability, daily snapshots since Sep 24.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via NVD
1.0%
Last analysed / modified upstream
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.
Refer to the linked advisories for vendor-supplied fixes and affected version ranges.
Connected by shared product, vendor, weakness, or advisory.
CVE-2026-61539Critical· 10.0Xinference is an inference API for running open-source, speech, and multimodal models
CVE-2026-112051Medium· 6.5Xinference through 4.0.1 Path Traversal via model_name in Custom Model Registration
CVE-2026-49950Critical· 9.8Xorbits Xoscar through 0.11.2 contains an unauthenticated deserialization vulnerability that allows remote attackers to execute arbitrary code by sending crafted data over the actor-pool TCP channel
CVE-2025-11905Medium· 6.3A vulnerability was found in yanyutao0402 ChanCMS up to 3.3.2
CVE-2025-14730Medium· 4.7A security flaw has been discovered in CTCMS Content Management System up to 2.1.2
CVE-2025-14729Medium· 4.7A vulnerability was identified in CTCMS Content Management System up to 2.1.2