Lucene search
+L

28 matches found

NVD
NVD
added 2026/08/24 2:17 p.m.6 views

CVE-2026-76841

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 trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

8.8CVSS0.0071EPSS
SaveExploits0References5
EUVD
EUVD
added 2026/08/24 1:11 p.m.7 views

EUVD-2026-64847

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 trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

8.7CVSS6.6AI score0.0071EPSS
SaveExploits0References5
Cvelist
Cvelist
added 2026/08/24 1:11 p.m.39 views

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

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 trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

8.8CVSS0.0071EPSS
SaveExploits0References5
CVE
CVE
added 2026/08/24 1:11 p.m.33 views

CVE-2026-76841

Xinference (by XorbitsAI) through version 2.11.0 unconditionally sets trust_remote_code=True at six model loader call sites spanning the rerank, embedding, and LLM modules. An attacker with model launch access can register a model with an unknown type and an arbitrary path, causing AutoTokenizer....

8.8CVSS6.5AI score0.0071EPSS
SaveExploits0References5
OSV
OSV
added 2026/08/24 1:11 p.m.17 views

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

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 trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

8.7CVSS6.8AI score
SaveExploits0References8
Vulnrichment
Vulnrichment
added 2026/08/24 1:11 p.m.7 views

CVE-2026-76841 Xinference through 2.11.0 Remote Code Execution via Hardcoded trust_remote_code in Model Loaders

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 trustremotecode=True as a literal or as an unconditional default: RerankModel.gettokenizer in...

8.8CVSS6.4AI score0.0071EPSS
SaveExploits0References5
NVD
NVD
added 2026/08/21 9:17 p.m.13 views

CVE-2026-61539

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

10CVSS0.00657EPSS
SaveExploits0References4
Github Security Blog
Github Security Blog
added 2026/08/21 8:56 p.m.42 views

Xinference vulnerable to remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Summary Xinference used Python's unsafe eval function when parsing Llama3 tool-call output generated by a large language model. Because the model output can be influenced by attacker-controlled prompts sent to the chat completion API, a remote attacker can craft prompts that cause the model to...

10CVSS6.5AI score0.00657EPSS
SaveExploits0References5Affected Software1
EUVD
EUVD
added 2026/08/21 8:37 p.m.12 views

EUVD-2026-64146

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

10CVSS5.5AI score0.00657EPSS
SaveExploits0References4
OSV
OSV
added 2026/08/21 8:37 p.m.46 views

CVE-2026-61539 Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

10CVSS5.8AI score
SaveExploits0References6
CVE
CVE
added 2026/08/21 8:37 p.m.75 views

CVE-2026-61539

Xinference (an inference API for open-source, speech, and multimodal models) versions 2.5.0 and earlier are vulnerable to remote code execution via unsafe use of Python's eval() in Llama3 tool-call parsing. An unauthenticated remote attacker can craft a prompt sent to /v1/chat/completions with a ...

10CVSS5.5AI score0.00657EPSS
SaveExploits0References4
Cvelist
Cvelist
added 2026/08/21 8:37 p.m.30 views

CVE-2026-61539 Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

10CVSS0.00657EPSS
SaveExploits0References4
Positive Technologies
Positive Technologies
added 2026/08/21 12:00 a.m.21 views

PT-2026-79457

Name of the Vulnerable Software and Affected Versions Xinference versions prior to 2.7.0 Description Xinference passes attacker-influenced Llama3 tool-call output to the eval function within xinference/model/llm/tool parsers/llama3 tool parser.py and xinference/model/llm/utils.py. Requests sent t...

10CVSS5.8AI score0.00657EPSS
SaveExploits0References10
OSSF Malicious Packages
OSSF Malicious Packages
added 2026/04/22 10:06 p.m.18 views

Malicious code in xinference (PyPI)

--- -= Per source details. Do not edit below this line.=- Source: kam193 1d006f6a08c959393160456d4ace221fd165b6d609fc8356ebfb041979aef93d Versions 2.6.0, 2.6.1, 2.6.2 were compromised. Following a malicious pull request that exfiltrated sensitive data from the CI runner, three malicious PyPI...

5.4AI score
SaveExploits0References6
OSV
OSV
added 2026/04/22 10:06 p.m.29 views

MAL-2026-3000 Malicious code in xinference (PyPI)

--- -= Per source details. Do not edit below this line.=- Source: kam193 1d006f6a08c959393160456d4ace221fd165b6d609fc8356ebfb041979aef93d Versions 2.6.0, 2.6.1, 2.6.2 were compromised. Following a malicious pull request that exfiltrated sensitive data from the CI runner, three malicious PyPI...

5.5AI score
SaveExploits0References6
Snyk
Snyk
added 2026/04/22 12:29 p.m.9 views

Embedded Malicious Code

Overview xinference is a powerful and versatile library designed to serve language, speech recognition, and multimodal models. With Xorbits Inference, you can effortlessly deploy and serve your or state-of-the-art built-in models using just a single command. Whether you are a researcher, develope...

9.8CVSS5.5AI score
SaveExploits0References2
EUVD
EUVD
added 2025/10/03 8:07 p.m.11 views

EUVD-2025-19750

Malicious code in bioql PyPI...

5.3CVSS6.5AI score0.00344EPSS
SaveExploits1References2
RedhatCVE
RedhatCVE
added 2025/07/04 12:23 a.m.20 views

CVE-2025-45424

Incorrect access control in Xinference before v1.4.0 allows attackers to access the Web GUI without authentication...

5.3CVSS7.2AI score0.00344EPSS
SaveExploits1References1
NVD
NVD
added 2025/07/02 5:15 p.m.11 views

CVE-2025-45424

Incorrect access control in Xinference before v1.4.0 allows attackers to access the Web GUI without authentication...

5.3CVSS0.00344EPSS
SaveExploits1References2
Snyk
Snyk
added 2025/07/02 4:42 p.m.5 views

Missing Authentication for Critical Function

Overview xinference is a Xorbits InferenceXinference is a powerful and versatile library designed to serve language, speech recognition, and multimodal models. With Xorbits Inference, you can effortlessly deploy and serve your or state-of-the-art built-in models using just a single command. Wheth...

7.3CVSS6.9AI score0.00344EPSS
SaveExploits1References2
Rows per page
Query Builder