---
id: CVE-2026-61539
title: >-
  Xinference is an inference API for running open-source, speech, and multimodal
  models
summary: >-
  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.…
severity: critical
cvss: 10
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H'
cwe:
  - CWE-95
vendor: xinference
product: xinference
affected:
  - xinference < 2.7.0
patched:
  - xinference 2.7.0
published: '2026-08-21'
updated: '2026-10-12'
sourceUpdated: '2026-10-12T16:10:00.220'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-61539'
references:
  - url: >-
      https://github.com/xorbitsai/inference/commit/1b3d220f342ce68d34cec4586d9409d457dadc42
    label: security-advisories@github.com
  - url: 'https://github.com/xorbitsai/inference/pull/4786'
    label: security-advisories@github.com
  - url: 'https://github.com/xorbitsai/inference/releases/tag/v2.7.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/xorbitsai/inference/security/advisories/GHSA-x2rj-828p-hx9m
    label: security-advisories@github.com
  - url: 'https://github.com/xorbitsai/inference'
  - url: 'https://pypi.org/project/xinference'
  - url: 'https://github.com/advisories/GHSA-x2rj-828p-hx9m'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-61539'
tags:
  - nvd
  - osv
  - pip
  - ghsa
epss: 0.01226
epssPercentile: 0.6891
aliases:
  - GHSA-x2rj-828p-hx9m
  - PYSEC-2026-3946
ecosystem: pip
ingestedAt: '2026-08-21T21:22:19.508Z'
---

## Overview

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.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-61539)

Affected packages:

- `xinference < 2.7.0`

Patched in:

- `xinference 2.7.0`

Source: https://osv.dev/vulnerability/GHSA-x2rj-828p-hx9m