Skip to content

Dataset · Mustafaege

SFT dataset for tool calling and Jupyter agents

Mustafaege/qwen3.5-toolcalling-v2

Over 60 thousand English examples for training models on external tools and code in Qwen3 format.

What's inside

Contains around 60 thousand SFT examples (55 thousand train and 6 thousand test) in English. The dataset combines data from UIGEN-X and code execution reasoning traces for Jupyter agents. All records are formatted to Qwen3 message format.

What you can do with it

  • SFT fine-tuning Qwen3 family models for API and external function calling
  • Training AI agents for code execution and reasoning in Jupyter
  • Testing tool-calling skills in language models
Tasks
Text generation
Languages
English
Size
100K–1M rows
Downloads this month
510
Likes
50

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Mustafaege/qwen3.5-toolcalling-v2")