Instructions to use Mindcraft-CE/Andy-4.20-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Mindcraft-CE/Andy-4.20-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Mindcraft-CE/Andy-4.20-GGUF",
	filename="andy-4.20.q8_0.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Mindcraft-CE/Andy-4.20-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Use Docker

docker model run hf.co/Mindcraft-CE/Andy-4.20-GGUF:Q8_0

LM Studio
Jan
Ollama
How to use Mindcraft-CE/Andy-4.20-GGUF with Ollama:
```
ollama run hf.co/Mindcraft-CE/Andy-4.20-GGUF:Q8_0
```

Unsloth Studio new

How to use Mindcraft-CE/Andy-4.20-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Mindcraft-CE/Andy-4.20-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Mindcraft-CE/Andy-4.20-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Mindcraft-CE/Andy-4.20-GGUF to start chatting

Pi new

How to use Mindcraft-CE/Andy-4.20-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Mindcraft-CE/Andy-4.20-GGUF:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Mindcraft-CE/Andy-4.20-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Run Hermes

hermes

Docker Model Runner
How to use Mindcraft-CE/Andy-4.20-GGUF with Docker Model Runner:
```
docker model run hf.co/Mindcraft-CE/Andy-4.20-GGUF:Q8_0
```

Lemonade

How to use Mindcraft-CE/Andy-4.20-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Mindcraft-CE/Andy-4.20-GGUF:Q8_0

Run and chat with the model

lemonade run user.Andy-4.20-GGUF-Q8_0

List all available models

lemonade list

The Mindcraft CE team introduces Andy-4.20, noted as the best local AI you can use to play Minecraft with. Thinking faster than Andy-4.1, being able to carry out more actions than Andy-4, and rivaling models 10x it's size.

Andy-4.20 is a specialized version of Andy-4.2, which uses XML tool calling for the experimental agent system instead of twitch commands.

Andy-4.20 will not work with any other version of Mindcraft-CE

Key Innovations

Andy-4.20 uses largely the same formula as Andy-4.1, but introduces a new architecture from the Qwen3.5 series which makes the model not only smarter, but more efficient. Using Gated Deltanet attention allows Andy-4.2 to run on a single RTX 3090, with 256k tokens of context, at a staggering 8-bit quantization.

Andy-4.20 is also the first local model capable of getting a full set of diamond armour, with zero human interaction

Like Andy-4.1, Andy-4.20 has vision capabilities, and has a stronger multimodal base that allows for even deeper comprehension of the game state.

How to Run

Andy-4.20 is still recommended to be ran using LM Studio, we have tried using Ollama, and there were a plethora of issues, including looping, mismatched chat templates, etc;

Below are the recommended sampling parameters for Andy-4.20, but the default settings in LM Studio work great, and the model is still able to get full diamond armour by itself.

Name	Value
Temperature	0.6
Repeat Penalty	1
Top P Sampling	0.95
Min P Sampling	0

Model Specifications

Size: 9B Parameters
Architecture: Qwen3.5
Context Length: Up to 1 million tokens
Message Count: 120 messages stable
CoT Style: DeepSeek-R1 style

Training Specifications

Hardware: 1x RTX 3090
Training Time: 5 Hours
Dataset Size: 2,748 examples
Learning Rate: 2e-5
LR Scheduler: cosine
Epoch Count: 1 Epoch
Training Quantization: 4-bit QLoRA with 8-bit QaT

Testing

The testing for Andy-4.20 was done at 8-bit, which was done to test if QaT (Quantization Aware Training) had assisted in the preservation of data inside of Andy-4.20.

Andy-4.2, which uses the same data as Andy-4.20, had the following stats during runtime for testing:

Mindcraft-CE version 1.2.7
8-bit Quantization
8-bit KV Cache quantization
Base LM Studio sampling parameters
32,000 Context Length
65 Messages

Limitations

Even though Andy-4.20 is capable of incredible feats, there is one domain where it does not perform well: Building. During internal testing any time Andy-4.2 would use !newAction, it would produce thousands and thousands of tokens, but never do anything. It is not advised to use Andy-4.2 as your code model.

Apart from that. Andy-4.20 has shown to be our most hard-working model yet, and navigates potential errors very well.

What's Next?

Based on the lessons from Andy-4.20, the Mindcraft team is prepared to collect better training data, explore new architectures to make the cost of running Andy models cheaper, as well as packing more brains into these tiny minds.

Licenses and Notices

Like all other Andy models, Andy-4.20 is based on the Andy license of terms. Being generally permissive, it contains qualifiers as to what makes an "Andy" class model.

See Andy 2.0 License.

This work uses data and models created by @Sweaterdog.

Downloads last month: 55

GGUF

Model size

9B params

Architecture

qwen35

Hardware compatibility

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Mindcraft-CE/Andy-4.20-GGUF

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Quantized

(216)

this model