Description
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/tool_parsers/llama3_tool_parser.py and xinference/model/llm/utils.py. Requests to /v1/chat/completions with a tools field flow through xinference/api/restful_api.py, xinference/model/llm/transformers/core.py, handle_chat_result_non_streaming(), and _post_process_completion() before extract_tool_calls() or _eval_llama3_chat_arguments() evaluates the model-generated Python expression. An unauthenticated remote attacker can influence that output through a crafted prompt and execute commands in the Xinference server process context. This issue is fixed in version 2.7.0.
Published: 2026-08-21
Score: 10 Critical
EPSS: < 1% Very Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Analysis and contextual insights are available on OpenCVE Cloud.

Remediation

No vendor fix or workaround currently provided.

Additional remediation guidance may be available on OpenCVE Cloud.

Tracking

Sign in to view the affected projects.

Advisories
Source ID Title
Github GHSA Github GHSA GHSA-x2rj-828p-hx9m Xinference vulnerable to remote code execution via unsafe `eval()` in Llama3 tool-call parsing
History

Mon, 24 Aug 2026 20:30:00 +0000

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'yes', 'Exploitation': 'none', 'Technical Impact': 'total'}, 'version': '2.0.3'}


Sun, 23 Aug 2026 16:45:00 +0000

Type Values Removed Values Added
First Time appeared Xorbitsai
Xorbitsai inference
Vendors & Products Xorbitsai
Xorbitsai inference

Fri, 21 Aug 2026 20:45:00 +0000

Type Values Removed Values Added
Description 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/tool_parsers/llama3_tool_parser.py and xinference/model/llm/utils.py. Requests to /v1/chat/completions with a tools field flow through xinference/api/restful_api.py, xinference/model/llm/transformers/core.py, handle_chat_result_non_streaming(), and _post_process_completion() before extract_tool_calls() or _eval_llama3_chat_arguments() evaluates the model-generated Python expression. An unauthenticated remote attacker can influence that output through a crafted prompt and execute commands in the Xinference server process context. This issue is fixed in version 2.7.0.
Title Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing
Weaknesses CWE-95
References
Metrics cvssV3_1

{'score': 10, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H'}


Subscriptions

Xorbitsai Inference
cve-icon MITRE

Status: PUBLISHED

Assigner: GitHub_M

Published:

Updated: 2026-08-24T19:45:13.875Z

Reserved: 2026-07-10T16:27:03.093Z

Link: CVE-2026-61539

cve-icon Vulnrichment

Updated: 2026-08-24T19:45:10.179Z

cve-icon NVD

Status : Received

Published: 2026-08-21T21:17:00.867

Modified: 2026-08-24T20:16:52.137

Link: CVE-2026-61539

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2026-08-23T16:40:35Z

Weaknesses