---
license: apache-2.0
language:
- en
base_model:
- unsloth/Ministral-3-3B-Instruct-2512
base_model_relation: adapter
library_name: peft
tags:
- canis-teach
- ministral
- education
- lora
- transformers
- tutoring
- math
pipeline_tag: text-generation
datasets:
- CanisAI/teach-math-v1
---

# Canis.teach - Ministral-3B Instruct (Math)

LoRA adapters for the Math tutor in the Canis.teach suite.

- **Base Model**: unsloth/Ministral-3-3B-Instruct-2512
- **Release**: CanisAI/teach-math-ministral-3b-r2
- **Project**: Canis.teach - Learning that fits.
- **Subject**: Math

## What is this?

This repository provides LoRA adapters fine-tuned on Math tutoring dialogues. Apply these adapters to the base model to enable subject-aware, didactic behavior without downloading a full merged checkpoint.

The model is designed to **teach, not just answer** - providing step-by-step explanations, hints, and pedagogically structured responses.

For ready-to-run merged models or Ollama-friendly GGUF quantizations, see the "Related Models" section.

## Quick Start

### Installation
```bash
pip install transformers peft torch
```

### Usage (LoRA)
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

base = "unsloth/Ministral-3-3B-Instruct-2512"
adapter = "CanisAI/teach-math-ministral-3b-r2"

tokenizer = AutoTokenizer.from_pretrained(base, use_fast=True)
model = AutoModelForCausalLM.from_pretrained(
    base, 
    device_map="auto",
    torch_dtype="auto"
)
model = PeftModel.from_pretrained(model, adapter)

# Example prompt
prompt = "Explain how to solve 2x + 1 = 5 step by step."
inputs = tokenizer.apply_chat_template(
    [{"role": "user", "content": prompt}],
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

outputs = model.generate(
    inputs,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.8,
    top_k=40,
    do_sample=True
)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

## Training Details

- **Base Model**: unsloth/Ministral-3-3B-Instruct-2512
- **Training Method**: Supervised Fine-Tuning (SFT) with LoRA
- **Framework**: Unsloth + TRL/PEFT
- **Data**: Canis.lab-curated Math tutoring dialogues
- **Target Modules**: Query, Key, Value, Output projections, MLP gates (gate, up, down)
- **Rank**: 32
- **Alpha**: 32

## Intended Use

- **Primary**: Subject-aware tutoring for Math education
- **Applications**: Educational prototypes, tutoring systems, research
- **Approach**: Stepwise explanations, pedagogical hints, rubric-aligned responses
- **Target Audience**: Students, educators, researchers

## Model Behavior

The model is optimized for:
- Clear, step-by-step explanations
- Appropriate difficulty progression  
- Encouraging learning through hints rather than direct answers
- Subject-specific pedagogical approaches
- Maintaining educational standards and accuracy

## Recommended Settings

For optimal tutoring behavior:
- **Temperature**: 0.6-0.8
- **Top-p**: 0.8-0.9
- **Top-k**: 20-40
- **Max tokens**: 512-1024

## Safety and Limitations

**Important Considerations**:
- Human oversight required for educational use
- May occasionally hallucinate or oversimplify complex topics
- For fact-critical applications, consider RAG with verified curriculum sources
- Follow your institution's data privacy and AI usage policies
- Not a replacement for qualified human instruction

## Related Models

| Type | Repository | Description |
|------|------------|-------------|
| **LoRA Adapters** | `CanisAI/teach-math-ministral-3b-r2` | This repository (lightweight) |
| **Merged Model** | (Coming Soon) | Ready-to-use full model | 
| **GGUF Quantized** | (Coming Soon) | Ollama/llama.cpp compatible | 
| **Dataset** | `CanisAI/teach-math-v1` | Training data |

## License

This model inherits the license from the base model (unsloth/Ministral-3-3B-Instruct-2512). Please review the base model's license terms before use.

## Citation

```bibtex
@misc{canis-teach-teach-math,
  title={Canis.teach Math Tutor},
  author={CanisAI},
  year={2026},
  publisher={Hugging Face},
  howpublished={\url{https://huggingface.co/CanisAI/teach-math-ministral-3b-r2}}
}
```

## Acknowledgments

- **MistralAI/Ministral Team** for the excellent base model
- **Unsloth** for efficient training tools
- **Hugging Face** ecosystem (Transformers, PEFT, TRL)
- Educators and contributors supporting the Canis.teach project

---
**Canis.teach** - Learning that fits.