Skip to content

Dataset · zgcagi

ZGCM-1 math and search dataset

zgcagi/ZGCM-1-Data

Large bilingual text dataset for mathematical reasoning and agentic search.

What's inside

This dataset supports the open ZGCM-1 model, a 7.39B-parameter dense model for mathematics and tool-assisted search. Metadata indicates text generation and question-answering formats in Chinese and English, with a size range of 1B to 10B tokens. The card provides links to the model, training logs, and code, though specific data composition is not detailed.

What you can do with it

  • Fine-tune LLMs for mathematical reasoning
  • Evaluate agentic search and question-answering
  • Build SFT datasets for bilingual math models
Tasks
Text generation, Question answering
Languages
Chinese, English
Size
1B–10B rows
Downloads this month
25K
Likes
97
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("zgcagi/ZGCM-1-Data")