---
license: apache-2.0
language:
- en
base_model: lumina-core
tags:
- bytarch
- belize
- education
- research
- deminji
- gguf
- bytarch
thumbnail: >-
  https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/model_card_template_default.png
pipeline_tag: text-generation
---

# 🇧🇿 BytArch-Lumina-0.6B

**BytArch-Lumina-0.6B** is a proprietary, high-velocity AI model developed by **BytArch LLC** in Belmopan, Belize. It serves as the intelligent core for the **Deminji Suite**, providing researchers, scholars, and developers with instant access to the BytArch ecosystem and API documentation.

Created and managed by **BytArch**, Lumina is optimized for speed, precision, and ethical academic research assistance.

## Model Variants
This repository contains one versions of the Lumina brain:
1.  **bytarch-lumina-0.6b.Q4_K_M.gguf**: The "Speed Copy." Compressed 4-bit version optimized for local use on laptops and mobile devices without sacrificing core logic.

## Specialized Knowledge Base
Lumina has been fine-tuned on the internal BytArch dataset, giving it expert knowledge in:
* **The Deminji Suite:** Tools for academic scholars and researchers.
* **Genie Platform:** Operational details for both Free and Premium tiers.
* **BytArch API:** Documentation, authentication, and endpoint management (https://api.bytarch.dpdns.org/openai/v1).
* **Regional Context:** Deep understanding of BytArch's mission as Belize’s premier AI cloud provider.

## Usage with Ollama

To run BytArch-Lumina locally, use the provided `Modelfile`.

### 1. Create the model
```bash
# For the standard version
ollama create bytarch-lumina -f Modelfile