{"id":"CVE-2026-76841","state":"PUBLISHED","title":"Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders","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.","publishedAt":"2026-08-24T13:11:59.094Z","updatedAt":"2026-10-01T15:21:12.439Z","severity":{"score":8.7,"label":"High","version":"4.0","vector":"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","source":"VulnCheck"},"assigner":"VulnCheck","products":["xorbitsai inference"],"affected":[{"vendor":"xorbitsai","product":"inference","packageName":"xinference","packageURL":"pkg:pypi/xinference","defaultStatus":"unaffected","versions":[{"version":"0","status":"affected","lessThan":"2.12.0","versionType":"semver"}],"collectionURL":"https://pypi.org/project/xinference/"}],"references":["https://github.com/xorbitsai/inference","https://github.com/xorbitsai/inference/issues/5023","https://github.com/xorbitsai/inference/pull/5027","https://github.com/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py","https://www.vulncheck.com/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders"],"weaknesses":["CWE-94"],"ingestedAt":"2026-10-29T20:12:58.252Z","source":{"repository":"CVEProject/cvelistV5","path":"cves/2026/76xxx/CVE-2026-76841.json","commit":"556c9d3a8cd17e218bb8e452ba7cd619a384ff47"},"raw":{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-76841","state":"PUBLISHED","datePublished":"2026-08-24T13:11:59.094Z","dateUpdated":"2026-10-01T15:21:12.439Z","assignerShortName":"VulnCheck","assignerOrgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateReserved":"2026-08-19T20:34:19.724Z"},"containers":{"cna":{"title":"Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders","descriptions":[{"lang":"en","value":"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":[{"vendor":"xorbitsai","product":"inference","packageName":"xinference","packageURL":"pkg:pypi/xinference","defaultStatus":"unaffected","versions":[{"version":"0","status":"affected","lessThan":"2.12.0","versionType":"semver"}],"collectionURL":"https://pypi.org/project/xinference/"}],"references":[{"url":"https://github.com/xorbitsai/inference","tags":["product"]},{"url":"https://github.com/xorbitsai/inference/issues/5023","tags":["issue-tracking"]},{"url":"https://github.com/xorbitsai/inference/pull/5027","tags":["issue-tracking","patch"]},{"url":"https://github.com/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py","tags":["technical-description"]},{"url":"https://www.vulncheck.com/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders","name":"VulnCheck Advisory: Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders","tags":["third-party-advisory"]}],"metrics":[{"format":"CVSS","cvssV4_0":{"version":"4.0","vectorString":"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","baseScore":8.7,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","attackRequirements":"NONE","privilegesRequired":"LOW","userInteraction":"NONE","vulnConfidentialityImpact":"HIGH","vulnIntegrityImpact":"HIGH","vulnAvailabilityImpact":"HIGH","subConfidentialityImpact":"NONE","subIntegrityImpact":"NONE","subAvailabilityImpact":"NONE"}},{"format":"CVSS","cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":8.8,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"LOW","userInteraction":"NONE","scope":"UNCHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-94","lang":"en","description":"Improper Control of Generation of Code ('Code Injection')","type":"CWE"}]}],"providerMetadata":{"shortName":"VulnCheck","orgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateUpdated":"2026-10-01T15:21:12.439Z"},"datePublic":"2026-08-01T00:00:00.000Z","cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:a:xinference:xinference:*:*:*:*:*:*:*:*","versionEndExcluding":"2.12.0"}]}]}],"credits":[{"lang":"en","value":"Fiona","type":"finder"}],"x_generator":{"engine":"vulncheck-endgame"}},"adp":[{"title":"CISA ADP Vulnrichment","metrics":[{"other":{"type":"ssvc","content":{"timestamp":"2026-08-26T15:56:50.906437Z","id":"CVE-2026-76841","options":[{"Exploitation":"none"},{"Automatable":"no"},{"Technical Impact":"total"}],"role":"CISA Coordinator","version":"2.0.3"}}}],"providerMetadata":{"shortName":"CISA-ADP","orgId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","dateUpdated":"2026-08-26T16:15:31.356Z"}}]}}}