This is a reasoning dataset generated using the stealth model Healer Alpha.
As the largest dataset we have made yet. The prompts from this dataset were almost all generated by Opus 4.5/4.6, GPT 5.1 and Gemini 3 (flash and pro).
The categories covered include academia, multi-lingual creative writing, finance, health, law, marketing/SEO, programming, philosophy, web dev, python scripting, and science.
Stats:
Cost: $ 0 (USD)
Tokens (input + output): 135 M
