CVE-2026-76841

Estado: AplazadaAlta (8.7)—

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.

CVSS

Versión: 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/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Puntuación base: 8.7

Probabilidad de explotación (EPSS)

Probabilidad de explotación en los próximos 30 días: 1.03%
Percentil entre todas las CVEs puntuadas: 63
Fecha de la puntuación: 6/10/2026

EPSS (Exploit Prediction Scoring System, de FIRST) estima la probabilidad de que una vulnerabilidad sea explotada en 30 días. Complementa a CVSS (impacto) y a CISA KEV (explotación confirmada).

🎯 Técnicas ATT&CK

Cómo se explota esta vulnerabilidad y qué consigue el atacante, en el lenguaje de MITRE ATT&CK.

ExplotaciónT1210 Exploitation of Remote Serviceslateral movement90 %
Impacto principalT1059 Command and Scripting Interpreterexecution95 %
Impacto secundarioT1068 Exploitation for Privilege Escalationprivilege escalation75 %

Acceso remoto requiere PR:L (privilegios de usuario autenticado). La vulnerabilidad permite ejecutar código Python arbitrario a través de trust_remote_code=True en modelos de Hugging Face, culminando en RCE con privilegios del worker.

Inferido por nuestro agente de análisis a partir de la descripción oficial, el vector CVSS y la CWE, y comprobado por un supervisor. Puede contener errores.

🛡️ Mitigaciones ATT&CK que cubren estas técnicas

Threat Intelligence Program Privileged Account Management Execution Prevention Disable or Remove Feature or Program Application Isolation and Sandboxing Exploit Protection

Tecnologías afectadas (1)

⚠ Inferidas por IA a partir de la descripción — NVD aún no ha analizado esta CVE; no son CPE verificados.

Xinference — Xinference^AI

CWE

CWE-94

Referencias

JSON original (NVD)

Mostrar

{
  "id": "CVE-2026-76841",
  "cveTags": [],
  "metrics": {
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-76841",
          "role": "CISA Coordinator",
          "options": [
            {
              "exploitation": "none"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "total"
            }
          ],
          "version": "2.0.3",
          "timestamp": "2026-08-26T15:56:50.906437Z"
        }
      }
    ],
    "cvssMetricV31": [
      {
        "type": "Secondary",
        "source": "disclosure@vulncheck.com",
        "cvssData": {
          "scope": "UNCHANGED",
          "version": "3.1",
          "baseScore": 8.8,
          "attackVector": "NETWORK",
          "baseSeverity": "HIGH",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
          "integrityImpact": "HIGH",
          "userInteraction": "NONE",
          "attackComplexity": "LOW",
          "availabilityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "confidentialityImpact": "HIGH"
        },
        "impactScore": 5.9,
        "exploitabilityScore": 2.8
      }
    ],
    "cvssMetricV40": [
      {
        "type": "Secondary",
        "source": "disclosure@vulncheck.com",
        "cvssData": {
          "Safety": "NOT_DEFINED",
          "version": "4.0",
          "Recovery": "NOT_DEFINED",
          "baseScore": 8.7,
          "Automatable": "NOT_DEFINED",
          "attackVector": "NETWORK",
          "baseSeverity": "HIGH",
          "valueDensity": "NOT_DEFINED",
          "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/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
          "exploitMaturity": "NOT_DEFINED",
          "providerUrgency": "NOT_DEFINED",
          "userInteraction": "NONE",
          "attackComplexity": "LOW",
          "attackRequirements": "NONE",
          "privilegesRequired": "LOW",
          "subIntegrityImpact": "NONE",
          "vulnIntegrityImpact": "HIGH",
          "integrityRequirement": "NOT_DEFINED",
          "modifiedAttackVector": "NOT_DEFINED",
          "subAvailabilityImpact": "NONE",
          "vulnAvailabilityImpact": "HIGH",
          "availabilityRequirement": "NOT_DEFINED",
          "modifiedUserInteraction": "NOT_DEFINED",
          "modifiedAttackComplexity": "NOT_DEFINED",
          "subConfidentialityImpact": "NONE",
          "vulnConfidentialityImpact": "HIGH",
          "confidentialityRequirement": "NOT_DEFINED",
          "modifiedAttackRequirements": "NOT_DEFINED",
          "modifiedPrivilegesRequired": "NOT_DEFINED",
          "modifiedSubIntegrityImpact": "NOT_DEFINED",
          "modifiedVulnIntegrityImpact": "NOT_DEFINED",
          "vulnerabilityResponseEffort": "NOT_DEFINED",
          "modifiedSubAvailabilityImpact": "NOT_DEFINED",
          "modifiedVulnAvailabilityImpact": "NOT_DEFINED",
          "modifiedSubConfidentialityImpact": "NOT_DEFINED",
          "modifiedVulnConfidentialityImpact": "NOT_DEFINED"
        }
      }
    ]
  },
  "affected": [
    {
      "source": "disclosure@vulncheck.com",
      "affectedData": [
        {
          "vendor": "xorbitsai",
          "product": "inference",
          "versions": [
            {
              "status": "affected",
              "version": "0",
              "lessThan": "2.12.0",
              "versionType": "semver"
            }
          ],
          "packageURL": "pkg:pypi/xinference",
          "packageName": "xinference",
          "collectionURL": "https://pypi.org/project/xinference/",
          "defaultStatus": "unaffected"
        }
      ]
    }
  ],
  "published": "2026-08-24T14:17:01.760",
  "references": [
    {
      "url": "https://github.com/xorbitsai/inference",
      "source": "disclosure@vulncheck.com"
    },
    {
      "url": "https://github.com/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py",
      "source": "disclosure@vulncheck.com"
    },
    {
      "url": "https://github.com/xorbitsai/inference/issues/5023",
      "source": "disclosure@vulncheck.com"
    },
    {
      "url": "https://github.com/xorbitsai/inference/pull/5027",
      "source": "disclosure@vulncheck.com"
    },
    {
      "url": "https://www.vulncheck.com/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders",
      "source": "disclosure@vulncheck.com"
    }
  ],
  "vulnStatus": "Deferred",
  "weaknesses": [
    {
      "type": "Secondary",
      "source": "disclosure@vulncheck.com",
      "description": [
        {
          "lang": "en",
          "value": "CWE-94"
        }
      ]
    }
  ],
  "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."
    }
  ],
  "lastModified": "2026-10-01T16:17:58.837",
  "sourceIdentifier": "disclosure@vulncheck.com"
}