Independent Research.ai

A loose collection of scientists creating insights on the inner workings of AI models. This is a home for publishing independent research without affiliation or organization. We are interested in the impact of AI on todays society and society's impact on AI. We create open science, with open models, open code and open datasets.

Intrigued? Questions? Collaboration?
- contact@independentresearch.ai
recent code and models
- LLM reasoning models via iwSFT (github)
- 32B reasoning model (iwSFT fine-tune of Qwen2.5)
research published by Independent Researchers
- 2025-07-17 Supervised Fine Tuning on Curated Data is Reinforcement Learning (and can be improved)