Andy‑4 is an 8 billion‑parameter specialist model tuned for Minecraft gameplay via the Mindcraft framework. Trained on a single RTX 3090 over three weeks, Andy‑4 delivers advanced reasoning, multi‑step planning, and robust in‑game decision‑making. The Current version of Andy-4 is Andy-4-0516, this was the date training finished.

⚠️ Certification:
Andy‑4 is not yet certified by the Mindcraft developers. Use in production at your own discretion.

If you are downloading on Huggingface, follow these directions!

DO NOT Use the `Use This Model` feature in Huggingface!

Andy-4 Huggingface Install Directions

Method One:

Select the model you would like to use
Download the Modelfile
Once downloaded, open Modelfile in a text editor, and change the FROM parameter from YOUR/PATH/HERE to the download location of the gguf file, this has to be exact!
When changed, save the file, and open command terminal
(Optional if CMD isn't opened via file explorer) Navigate to the correct directory using "cd"
Run the command ollama create sweaterdog/Andy-4 -f Modelfile If you want multiple models, include a tag afterwards. Example: sweaterdog/Andy-4:micro-fp16 or sweaterdog/Andy-4:q2_k
Go to a profile in MindCraft
Change the model to be sweaterdog/Andy-4 Or whatever you named your model
Ensure you have the emdedding tag set to Ollama, like below

{
  "name": "andy-4",

  "model": "Sweaterdog/Andy-4",

  "embedding": "ollama"

}

Method Two:

Download the Modelfile
Once downloaded, open Modelfile in a text editor, and change the FROM parameter from YOUR/PATH/HERE To one of the models listed here in the Use This Model tab under ollama, here are the options: ```

hf.co/Sweaterdog/Andy-4:Q2_K

hf.co/Sweaterdog/Andy-4:Q3_K_M

hf.co/Sweaterdog/Andy-4:Q4_K_M

hf.co/Sweaterdog/Andy-4:Q5_K_M

hf.co/Sweaterdog/Andy-4:Q8_0

hf.co/Sweaterdog/Andy-4:F16

3. When changed, save the file, and open command terminal

(Optional if CMD isn't opened via file explorer) Navigate to the correct directory using "cd"
Run the command ollama create sweaterdog/Andy-4 -f Modelfile If you want multiple models, include a tag afterwards. Example: sweaterdog/Andy-4:micro-fp16 or sweaterdog/Andy-4:q2_k
Go to a profile in MindCraft
Change the model to be sweaterdog/Andy-4 Or whatever you named your model
Ensure you have the emdedding tag set to Ollama, like below

{
  "name": "andy-4",

  "model": "Sweaterdog/Andy-4",

  "embedding": "ollama"

}

DO NOT SKIP THIS SECTION IF YOU INTEND ON INSTALLING OFF OF HUGGINGFACE

🔍 Model Specifications

Parameters: 8 B
Training Hardware: 1 × NVIDIA RTX 3090
Duration: ~3 weeks total
Data Volumes:
- Messages: 179,384
- Tokens: 425,535,198
- Conversations: 62,149
Base Architecture: Deepseek-R1-LLaMA
License: Andy 1.0 License
Repository: https://huggingface.co/Sweaterdog/Andy‑4

📊 Training Regimen

Andy‑4‑base‑1 dataset
- Epochs: 3
- Learning Rate: 7e-5
- Dataset Size: 47.4k
Fine‑tune (FT) dataset
- Epochs: 2.5
- Learning Rate: 2e-5
- Dataset Size: 4.12k

Optimizer: AdamW_8bit with cosine decay
Quantization: 4‑bit (bnb-4bit) for inference
Warm Up Steps: 0.1% of each dataset

🚀 Installation

First, you need to choose your quantization, this chart is with the base of 8192 set as the context window

Quantization	VRAM Required
F16	16 GB+
Q5_K_M	8 GB+
Q4_K_M	6–8 GB
Q3_K_M	6 GB (low)
Q2_K	4–6 GB (ultra low)

1. Installation directly on Ollama

Visit Andy-4 on Ollama
Copy the command after choosing model type / quantization
Run the command in the terminal
Set the profile's model to be what you installed, such as ollama/sweaterdog/andy-4:latest

2. Manual Download & Modelfile

Download
- From the HF Files tab, grab your chosen .GGUF quant weights (e.g. Andy-4.Q4_K_M.gguf).
- Download the provided Modelfile.
Edit

Change

FROM YOUR/PATH/HERE

to

FROM /path/to/Andy-4.Q4_K_M.gguf

Optional: Increase the parameter num_ctx to a higher value for longer conversations if you:

A. Have extra VRAM

B. Quantized the context window

C. Can use a smaller model

Create
ollama create andy-4 -f Modelfile

This registers the Andy‑4 model locally.

If you lack a GPU, check the Mindcraft Discord guide for free cloud setups.

🔧 Context‑Window Quantization

To lower VRAM use for context windows:

Windows

Close Ollama.
In System Properties → Environment Variables, add:
text
1OLLAMA_FLASH_ATTENTION=1 2OLLAMA_KV_CACHE_TYPE=q8_0 # or q4_0 for extra savings, but far more unstable
Restart Ollama.

Linux/macOS

bash

export OLLAMA_FLASH_ATTENTION=1
export OLLAMA_KV_CACHE_TYPE="q8_0"   # or "q4_0", but far more unstable
ollama serve

📌 Acknowledgments

Click to expand

Data & Models by: @Sweaterdog
Framework: Mindcraft (https://github.com/kolbytn/mindcraft)
LoRA Weights: https://huggingface.co/Sweaterdog/Andy-4-LoRA
*Explicit credit is not granted to Meta since this model was trained off of a slightly different architecture, from DeepSeek-R1

⚖️ License

See Andy 1.0 License.

This work uses data and models created by @Sweaterdog.

Sweaterdog/Andy-4-0516-Safetensors

🧠 Andy‑4 ⛏️