| Reporter | Title | Published | Views | Family All 7 |
|---|---|---|---|---|
 | CVE-2026-76841 | 25 Aug 202605:20 | – | circl |
 | CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders | 24 Aug 202613:11 | – | cvelist |
 | EUVD-2026-64847 | 24 Aug 202613:11 | – | euvd |
 | CVE-2026-76841 | 24 Aug 202614:17 | – | nvd |
 | PT-2026-80851 | 24 Aug 202600:00 | – | ptsecurity |
 | CVE-2026-76841: Improper Control of Generation of Code | 24 Aug 202600:00 | – | rapid7 |
 | CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders | 24 Aug 202613:11 | – | vulnrichment |
[
{
"vendor": "xorbitsai",
"product": "inference",
"collectionURL": "https://pypi.org/project/xinference/",
"packageName": "xinference",
"packageURL": "pkg:pypi/xinference",
"defaultStatus": "unaffected",
"versions": [
{
"version": "0",
"lessThan": "2.12.0",
"status": "affected",
"versionType": "semver"
}
]
}
]| Parameter | Position | Path | Description | CWE |
|---|---|---|---|---|
| model_path | path | xinference/model/rerank/core.py | Unconditionally enables remote code execution by loading models with trust_remote_code, leading to attacker-supplied code execution during tokenizer loading via AutoTokenizer.from_pretrained. | CWE-94 |
| trust_remote_code | path | xinference/model/rerank/core.py | Unconditionally enables remote code execution by loading models with trust_remote_code, leading to attacker-supplied code execution during tokenizer loading via AutoTokenizer.from_pretrained. | CWE-94 |
| model_path | path | xinference/model/rerank/sentence_transformers/core.py | Rerank SentenceTransformer loader uses default trust_remote_code which can execute attacker-provided code during model/tokenizer loading. | CWE-94 |
| trust_remote_code | path | xinference/model/rerank/sentence_transformers/core.py | Rerank SentenceTransformer loader uses default trust_remote_code which can execute attacker-provided code during model/tokenizer loading. | CWE-94 |
| model_path | path | xinference/model/embedding/sentence_transformers/core.py | Embedding SentenceTransformer loader uses default trust_remote_code enabling remote code execution via tokenizer loading. | CWE-94 |
| trust_remote_code | path | xinference/model/embedding/sentence_transformers/core.py | Embedding SentenceTransformer loader uses default trust_remote_code enabling remote code execution via tokenizer loading. | CWE-94 |
| model_path | path | xinference/model/embedding/flag/core.py | FlagEmbedding loader uses default trust_remote_code enabling attacker-supplied code during tokenizer loading. | CWE-94 |
| trust_remote_code | path | xinference/model/embedding/flag/core.py | FlagEmbedding loader uses default trust_remote_code enabling attacker-supplied code during tokenizer loading. | CWE-94 |
| model_path | path | xinference/model/llm/transformers/core.py | PytorchModel._sanitize_model_config defaults to True, enabling remote code execution during model loading. | CWE-94 |
| trust_remote_code | path | xinference/model/llm/transformers/core.py | PytorchModel._sanitize_model_config defaults to True, enabling remote code execution during model loading. | CWE-94 |
10
Data
Build on a solid foundation with Vulners data
We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data
Api
Power your application with Vulners API
The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access
App
Assess and manage vulnerabilities with Vulners tools
Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation
01 Oct 2026 16:17Current
6.2Medium risk
Vulners AI Score6.2
CVSS 48.7
CVSS 3.18.8
EPSS0.01029
SSVC