Abc8264/TutorAI-Chemistry-Phi4

TutorAI: Chemistry Reasoning Model (Phi-4-Mini) 🧪

This is a specialized, 4-billion parameter Chemistry tutoring model fine-tuned using Group Relative Policy Optimization (GRPO). It is designed to act as an expert Chemistry Professor, breaking down complex molecular interactions, stoichiometry, and formulas step-by-step before providing a final answer.

This repository contains the heavily compressed Q4_K_M GGUF version of the model, optimized for fast, local inference on standard laptops (Mac/Windows/Linux) using tools like Ollama or LM Studio.

🧠 Model Capabilities & "Thinking" Format

Unlike standard conversational models, this model was trained via Reinforcement Learning to "show its work." It uses specialized XML tags to map out its chemical reasoning before outputting the final result.

When prompted correctly, the model will output:

1. <think> ... </think>: A detailed, step-by-step breakdown of the chemical concepts, formulas, or math required.
2. \boxed{...}: The final, definitive answer isolated for easy reading.

🚀 How to Run Locally

You can run this model instantly on your own machine using Ollama.

Option 1: Fast Start (Ollama CLI)

Run this command in your terminal to pull and run the model directly from this repository:

ollama run hf.co/YOUR_USERNAME/TutorAI-Chemistry-Phi4-GGUF