{
  "affected": [
    {
      "ranges": [
        {
          "database_specific": {
            "extracted_events": [
              {
                "introduced": "0"
              },
              {
                "fixed": "2.12.0"
              }
            ],
            "source": [
              "AFFECTED_FIELD",
              "DESCRIPTION"
            ]
          },
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0a310a515ce401e38937a58b7210b1fa874ee156"
            }
          ],
          "repo": "https://github.com/xorbitsai/inference",
          "type": "GIT"
        }
      ]
    }
  ],
  "database_specific": {
    "cna_assigner": "VulnCheck",
    "cwe_ids": [
      "CWE-94"
    ],
    "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/76xxx/CVE-2026-76841.json"
  },
  "details": "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.",
  "id": "CVE-2026-76841",
  "modified": "2026-08-30T03:30:32.813269707Z",
  "published": "2026-08-24T13:11:59.094Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/76xxx/CVE-2026-76841.json"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-76841"
    },
    {
      "type": "ADVISORY",
      "url": "https://www.vulncheck.com/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders"
    },
    {
      "type": "REPORT",
      "url": "https://github.com/xorbitsai/inference/issues/5023"
    },
    {
      "type": "FIX",
      "url": "https://github.com/xorbitsai/inference/pull/5027"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/xorbitsai/inference"
    },
    {
      "type": "PACKAGE",
      "url": "https://pypi.org/project/xinference/"
    },
    {
      "type": "ARTICLE",
      "url": "https://github.com/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py"
    }
  ],
  "schema_version": "1.9.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders"
}