Skip to content
Heist/ feed
Back to feed
High

Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

Type: advisorySources: OSV

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.

Products
Not specified
Affected versions
Not specified
Fixed versions
Not specified
CVEs
CVE-2026-76841
Confirmed fixed CVEs
Not specified
Updated in this feed
2 Nov 2026, 16:37 UTC · Revision 1

Source evidence

Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

core-osv · Original source

Published 24 Aug 2026, 13:11 UTC

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.

osv:CVE-2026-76841
Complete source record
{
  "id": "25b3a97120774f10a995622111c2c77816c57de1785bf59cb262f6602c560ff2",
  "url": "https://osv.dev/vulnerability/CVE-2026-76841",
  "method": "structured",
  "payload": {
    "kind": "advisory",
    "title": "Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders",
    "summary": "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.",
    "evidence": "osv:CVE-2026-76841",
    "products": [],
    "severity": "high",
    "structured": {
      "id": "CVE-2026-76841",
      "aliases": [
        "CVE-2026-76841"
      ],
      "metrics": [
        {
          "type": "CVSS_V4",
          "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"
        }
      ],
      "affected": [
        {
          "ranges": [
            {
              "repo": "https://github.com/xorbitsai/inference",
              "type": "GIT",
              "events": [
                {
                  "introduced": "0"
                },
                {
                  "fixed": "0a310a515ce401e38937a58b7210b1fa874ee156"
                }
              ],
              "database_specific": {
                "extracted_events": [
                  {
                    "range": [
                      {
                        "introduced": "0"
                      },
                      {
                        "fixed": "2.12.0"
                      }
                    ],
                    "source": "AFFECTED_FIELD",
                    "strategy": "StandardRange"
                  },
                  {
                    "range": [
                      {
                        "introduced": "0"
                      },
                      {
                        "fixed": "2.12.0"
                      }
                    ],
                    "source": "DESCRIPTION"
                  }
                ]
              }
            }
          ],
          "versions": [
            "v2.11.0",
            "v2.10.0",
            "v2.9.0",
            "v2.8.0",
            "v2.7.0",
            "v2.5.0",
            "v2.4.0",
            "v2.3.0",
            "v2.2.0",
            "v2.1.0",
            "v2.0.0",
            "v1.17.0",
            "v1.16.0",
            "v1.15.0",
            "v1.14.0",
            "v1.13.0",
            "v1.12.0",
            "v1.11.0.post1",
            "v1.11.0",
            "v1.10.1",
            "v1.10.0",
            "v1.9.1",
            "v1.9.0",
            "v1.8.1.rc1",
            "v1.8.1",
            "v1.8.0",
            "v1.7.1.post1",
            "v1.7.1",
            "v1.7.0.post1",
            "v1.7.0",
            "v1.6.1",
            "v1.6.0.post1",
            "v1.6.0",
            "v1.5.1",
            "v1.5.0.post1",
            "v1.5.0",
            "v1.4.1",
            "v1.4.0",
            "v1.3.1.post1",
            "v1.3.1",
            "v1.3.0.post2",
            "v1.3.0.post1",
            "v1.2.2",
            "v1.2.1",
            "v1.2.0",
            "v1.1.1",
            "v1.1.0",
            "v1.0.1",
            "v1.0.0",
            "v0.16.3",
            "v0.16.2",
            "v0.16.1",
            "v0.16.0",
            "v0.15.4",
            "v0.15.3",
            "v0.15.2",
            "v0.15.1",
            "v0.15.0",
            "v0.14.4.post1",
            "v0.14.4",
            "v0.14.3",
            "v0.14.2",
            "v0.14.1.post1",
            "v0.14.1",
            "v0.14.0.post1",
            "v0.14.0",
            "v0.13.3",
            "v0.13.2",
            "v0.13.1",
            "v0.13.0",
            "v0.12.3",
            "v0.12.2.post1",
            "v0.12.2",
            "v0.12.1",
            "v0.12.0",
            "v0.11.3",
            "v0.11.2.post1",
            "v0.11.2",
            "v0.11.1",
            "v0.11.0",
            "v0.10.3",
            "v0.10.2.post1",
            "v0.10.2",
            "v0.10.1",
            "v0.10.0",
            "v0.9.4",
            "v0.9.3",
            "v0.9.2",
            "v0.9.1",
            "v0.9.0",
            "v0.8.5",
            "v0.8.4",
            "v0.8.3.1",
            "v0.8.3",
            "v0.8.2",
            "v0.8.1",
            "v0.8.0",
            "v0.7.5",
            "v0.7.4.1",
            "v0.7.4",
            "v0.7.3.1",
            "v0.7.3",
            "v0.7.2",
            "v0.7.1",
            "v0.7.0",
            "v0.6.5",
            "v0.6.4",
            "v0.6.3",
            "v0.6.2",
            "v0.6.1",
            "v0.6.0",
            "v0.5.6",
            "v0.5.5",
            "v0.5.4",
            "v0.5.3",
            "v0.5.2",
            "v0.5.1",
            "v0.5.0",
            "v0.4.4",
            "v0.4.3",
            "v0.4.2",
            "v0.4.1",
            "v0.4.0",
            "v0.3.0",
            "v0.2.3",
            "v0.2.2",
            "v0.2.1",
            "v0.2.0",
            "v0.1.3",
            "v0.1.2",
            "v0.1.1",
            "v0.1.0",
            "v0.0.6",
            "v0.0.5",
            "v0.0.4",
            "v0.0.3",
            "v0.0.2",
            "v0.0.1"
          ],
          "database_specific": {
            "source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-76841.json"
          }
        }
      ],
      "modified": "2026-10-28T02:30:21.090468200Z",
      "provider": "osv",
      "references": [
        "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/76xxx/CVE-2026-76841.json",
        "https://nvd.nist.gov/vuln/detail/CVE-2026-76841",
        "https://www.vulncheck.com/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders",
        "https://github.com/xorbitsai/inference/issues/5023",
        "https://github.com/xorbitsai/inference/pull/5027",
        "https://github.com/xorbitsai/inference",
        "https://pypi.org/project/xinference/",
        "https://github.com/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py"
      ],
      "provider_metadata": {
        "cwe_ids": [
          "CWE-94"
        ],
        "cna_assigner": "VulnCheck",
        "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/76xxx/CVE-2026-76841.json"
      }
    },
    "related_cves": [
      {
        "value": "CVE-2026-76841",
        "evidence": "/id"
      }
    ],
    "security_relevant": true,
    "severity_evidence": "/severity/0/score"
  },
  "source_id": "core-osv",
  "document_id": 487247,
  "published_at": "2026-08-24T13:11:59.094+00:00"
}