Llama-trina-mc-3b

Llama-trina-mc-3b is a custom, fine-tuned conversational AI model designed specifically to act as an intelligent assistant for Minecraft servers.

Model Overview

  • Base Model: Llama 3.2
  • Parameters: 3B
  • Focus: Minecraft mechanics, deep lore, server administration, and casual talkative chat.
  • Format: MLX (optimized for Apple Silicon) and standard Safetensors.

Training Data

The model was fine-tuned on a custom dataset containing:

  • Over 1,600 unique Minecraft crafting recipes.
  • In-depth game mechanics (redstone, mob farms, villager trading, etc.).
  • A custom talkative persona designed to act as a friendly and helpful server companion named Trina.

Usage

You can run this model natively on Mac using MLX:

pip install mlx-lm
mlx_lm.generate --model yucchun/Llama-trina-mc-3b --prompt "How do I build an iron golem farm?"

Plugin Integration

This model was specifically trained to be used with the Trina Minecraft Server Plugin. You can find the open-source server plugin code here: heyotw/trina on GitHub

License Compliance

This model is built upon Meta's Llama 3.2 materials. Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

By using this model, you agree to the Llama 3.2 Community License Agreement and the Llama 3.2 Acceptable Use Policy.

Downloads last month
6
Safetensors
Model size
3B params
Tensor type
BF16
·
MLX
Hardware compatibility
Log In to add your hardware

Quantized

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