CVE-2026-76841 json
Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders
published 2mo ago modified 30d ago also EUVD-2026-64847
- cvss
- 8.8
- version 3.1, CNA
- cisa kev
- Not listed
- not in CISA's catalog
- epss
- 1.0%
- EPSS 1.0% chance of exploitation in 30 days, percentile 64%
- public exploits
- 0
- none known
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.
Patch / fix
Is CVE-2026-76841 in your stack? Fencer shows which of your repositories are affected as soon as an advisory lands, and opens the fix as a pull request.
Check with FencerCISA SSVC
- Exploitation
- none
- Automatable
- no
- Technical impact
- total
- Assessed
- 2mo ago
Exploit prediction
1.03% chance of exploitation in the next 30 days, higher than 64% of all scored CVEs (FIRST EPSS, 2026-11-11).
CVSS scores
| assigned by | version | score | vector |
|---|---|---|---|
| CNA | 4.0 | 8.7 | 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 |
| CNA | 3.1 | 8.8 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H |
| CNA | 4.0 | 8.7 | 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 |
Affected products
- xorbitsai inference
Classification
- Weakness
-
- CWE-94 Improper Control of Generation of Code
References
-
github.com
/xorbitsai/inference/pull/5027
issue-tracking, patch
-
github.com
/xorbitsai/inference
product
-
github.com
/xorbitsai/inference/issues/5023
issue-tracking
-
github.com
/xorbitsai/inference/blob/v2.11.0/xinference/model/rerank/core.py
technical-description
-
vulncheck.com
/advisories/xinference-through-remote-code-execution-via-hardcoded-trust-remote-code-in-model-loaders
third-party-advisory
Sources
- CVE List CVE-2026-76841
- NVD CVE-2026-76841
Fencer Intel keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.
CPEs and fixed versions use Anchore's correction of NVD's data: Added CPE configurations because not yet analyzed by NVD.