Goal Reached Thanks to every supporter — we hit 100%!

Goal: 1000 CNY · Raised: 1336 CNY

100%
Buy Us a Coffee

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

Quick assessment

Affected
xorbitsai inference
Exploitation
Public or AI PoC available; prioritize validation
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.

Xinference 是一个用于运行开源大语言模型、语音模型和多模态模型的推理 API。在 2.5.0 及更早版本中，Xinference 将受攻击者影响的 Llama3 工具调用输出传递给了 和 中的 函数。带有 字段的 请求会经过 、 、 和 ，最终在 或 中对模型生成的 Python 表达式进行评估。未授权的远程攻击者可通过构造特定提示（prompt）影响该输出，从而在 Xinference 服务器进程上下文中执行任意命令。该漏洞已在 2.7.0 版本中修复。

CVSS 10.0 · Critical EPSS 0.66% · P48

Possible ATT&CK Techniques 1 AI

T1059 · Command and Scripting Interpreter

Affected Version Matrix 1

VendorProduct Version RangeStatus
xorbitsai inference < 2.7.0 affected
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-61539

Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗

Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.

Vulnerability Title
Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing
Source: CVE Program / CVE List V5
Vulnerability 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.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
Source: CVE Program / CVE List V5
Vulnerability Type
动态执行代码中指令转义处理不恰当(Eval注入)
Source: CVE Program / CVE List V5

Affected Products

Vendor Product Affected Versions CPE Subscribe
xorbitsai inference < 2.7.0 -

II. Public POCs for CVE-2026-61539

# POC Description Source Link Shenlong Link
AI-Generated POC Premium
Qwen3.6-35B-A3B · 6093 chars
Pro+ exclusive includes:
Vulnerability reproduction recording (real sandbox build + trigger, exclusive)
In-depth vulnerability mechanism
Trigger conditions & impact
Full executable POC code
Exploit chain & mitigation
POC zip download
100+ AI POC generations per month

III. Intelligence Information for CVE-2026-61539

请登录查看更多情报信息。

Patches & Fixes for CVE-2026-61539 (2)

Vendor Advisories for CVE-2026-61539 (1)

Vendor Pages for CVE-2026-61539 (1)

IV. Related Vulnerabilities

Same product: inference
Same weakness: CWE-95

V. Comments for CVE-2026-61539

No comments yet

Leave a comment