Tether-backed QVAC Launches Genesis II Dataset

QVAC Unveils Genesis II, Expanding AI Dataset to 148 Billion Tokens

• QVAC, Tether Data’s AI division, released Genesis II, adding 107 billion tokens to its synthetic dataset for AI pre-training.
• The dataset now totals 148 billion tokens across 19 education-focused domains, making it the largest public educational synthetic dataset.
• Genesis II includes new fields such as computer science and machine learning and introduces an “Option-Level Reasoning” approach.
• Independent evaluations show improved reasoning accuracy and clarity in models trained on Genesis II data compared to previous sets.
• The dataset is available under a Creative Commons license on QVAC’s blog and Hugging Face, supporting open research and local model development.

Tether Data’s QVAC has significantly expanded its synthetic educational dataset with Genesis II, now encompassing a total of 148 billion tokens. This initiative enhances AI model capabilities through structured understanding and broader domain coverage.