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

Severity: high (CVSS 8.8 v3.1)
Exploited in the wild: not listed in CISA KEV
EPSS: 1.0% probability of exploitation in the next 30 days
Public exploit or detection template: none known
Published: 2026-08-24 · updated 2026-10-12
Weakness: CWE-94

Description

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.

Affected products

xorbitsai inference: before 2.12.0 — fixed in 2.12.0

How to fix

Upgrade to a fixed version: xorbitsai inference 2.12.0. Check the vendor advisory for mitigations.

References

Get alerted when CVE-2026-76841 or the next vulnerability affects your stack