Skip to content

Dataset · PolarSeeker

Training data for AI search agents

PolarSeeker/OpenSeeker-v1-Data

An open English corpus for training agents capable of information retrieval and question answering.

What's inside

The dataset contains between 10K and 100K English text examples for the question-answering task (the description mentions a sample of 11.7K training examples). The authors fine-tuned the Qwen3-30B-A3B-Thinking-2507 model on this data to operate as a search agent.

What you can do with it

  • Training and SFT tuning of search AI agents
  • Developing thinking skills when searching for answers to complex questions
  • Evaluating agent performance on information retrieval tasks
Tasks
Question answering
Languages
English
Size
10K–100K rows
Downloads this month
1.4K
Likes
55

How to load it

from datasets import load_dataset

ds = load_dataset("PolarSeeker/OpenSeeker-v1-Data")