EUVD-2026-64847
GHSA-9cx2-fwcj-qq73
Severity by source
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
Network-accessible with low-privilege model-launch credential required (PR:L, AV:N, AC:L); no scope change as impact is confined to the worker process (S:U); full C/I/A impact via arbitrary code execution.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
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
Lifecycle Timeline
4DescriptionCVE.org
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.
Articles & Coverage 1
AnalysisAI
Remote code execution in Xinference (all versions before 2.12.0) allows any user with model-launch access to execute arbitrary Python code on the server at worker-process privilege level. Six model loader call sites across the rerank, embedding, and LLM subsystems hardcode trust_remote_code=True when invoking Hugging Face AutoTokenizer.from_pretrained, meaning a registered model with an attacker-controlled path can ship a malicious tokenizer_config.json whose auto_map entries are imported and executed unconditionally upon model load. …
Unlock full vulnerability intelligence
- Risk assessment & exploitation conditions
- Attack chain visualization
- Remediation with exact patch versions
- Threat intelligence from 22 sources
- Personal watchlist & email alerts
Free forever · No credit card required
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata
Vulnerability AssessmentAI
| Exploitation | Exploitation requires a user account with model-launch (model registration) access on the Xinference server, consistent with CVSS PR:L - fully unauthenticated exploitation is not possible. … Additional conditions and limiting factors are described in the full assessment. |
| Risk Assessment | The provided CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N, score 8.7) reflects network-accessible exploitation requiring only low-privilege authentication, with high impact on confidentiality, integrity, and availability of the vulnerable system - an accurate characterization of the attack surface. … Full risk analysis with EPSS, KEV, and SSVC signal comparison available after sign-in. |
| Exploit Scenario | An attacker with a valid Xinference account and model-launch permission constructs a malicious directory containing a tokenizer_config.json with an auto_map entry that references an attacker-supplied Python file containing arbitrary OS commands or a reverse shell payload. The attacker registers this model via the Xinference API supplying the malicious directory as the model path, then triggers model loading; AutoTokenizer.from_pretrained unconditionally imports the attacker's Python module, executing the payload at the privilege level of the Xinference worker process. … |
| Remediation | The primary fix is to upgrade to Xinference version 2.12.0, which introduces the allow_trust_remote_code utility function and gates all six previously unconditional loader call sites behind it. … Detailed patch versions, workarounds, and compensating controls in full report. |
Recommended ActionAI
Within 24 hours: audit inventory for Xinference deployments and determine current versions and whether user-defined model registration is enabled. …
Sign in for detailed remediation steps and compensating controls.
Threat intelligence, references, and detailed analysis are available after sign-in.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
Cross-user flow execution in Langflow (< 1.9.1) lets any authenticated API-key holder run another user's flow by passing
Same weakness CWE-94 – Code Injection
View allSame technique Code Injection
View allShare
External POC / Exploit Code
Leaving vuln.today