Skip to content
Fencer Intel
high 8.8

CVE-2025-9141 json

vLLM has remote code execution vulnerability in the tool call parser for Qwen3-Coder

published last yr. modified last mo. also GHSA-79j6-g2m3-jgfw , PYSEC-2026-2014

cvss
8.8
version 3.1, GitHub
cisa kev
Not listed
not in CISA's catalog
epss
—
no EPSS score yet
public exploits
0
none known

Summary

### Summary An unsafe deserialization vulnerability allows any authenticated user to execute arbitrary code on the server if they are able to get the model to pass the code as an argument to a tool call.

### Details vLLM's [Qwen3 Coder tool parser](https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/openai/tool_parsers/qwen3coder_tool_parser.py) contains a code execution path that uses Python's `eval()` function to parse tool call parameters. This occurs during the parameter conversion process when the parser attempts to handle unknown data types.

This code path is reached when: 1. Tool calling is enabled (`--enable-auto-tool-choice`) 2. The qwen3_coder parser is specified (`--tool-call-parser qwen3_coder`) 3. The parameter type is not explicitly defined or recognized

### Impact Remote Code Execution via Python's `eval()` function.

Patch / fix

  • pypi vllm 0.10.1.1

Is CVE-2025-9141 in your dependencies? Fencer shows which of your repositories are affected as soon as an advisory lands, and opens the fix as a pull request.

Check with Fencer

Vendor advisories

What the vendors themselves say about their products: affected, fixed or not affected, and what to do.

red hat product security vex CVE-2025-9141 updated cvss 8.8

vllm: quen3: RCE in vllm tool call parser for qwen3coder

  • Red Hat Enterprise Linux AI (RHEL AI) rhelai1/smollm3-3b, rhelai1/gemma-3n-e4b-it, rhelai1/modelcar-smollm3-3b, rhelai1/modelcar-gemma-3n-e4b-it, rhelai1/qwen3-4b-quantized-w4a16, rhelai1/deepseek-r1-0528-quantized-w4a16, +4 more
    not affected: vulnerable code not present

Red Hat CSAF VEX © Red Hat, Inc., from https://security.access.redhat.com/data/csaf/v2/vex/; changed: reduced to product statuses and remediations. CC BY 4.0 · csaf

CVSS scores

assigned by version score vector
GitHub 3.1 8.8 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
OSV 3.1 8.8 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Affected packages

ecosystem package affected fixed in
pypi vllm
>=0.10.0, <0.10.1.1
0.10.1.1

Classification

Weakness
  • CWE-502 Deserialization of Untrusted Data

References

Sources

Fencer Intel keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.