Downloads · 30 days
3
7% of all-time downloads
Luna-2.5B-LLM
Arunachallam/Luna-2.5B-LLM
Luna-2.5B-LLM is a machine learning model from Arunachallam. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Luna-2.5B is a customized and optimized derivative of the open-source Qwen2.5-1.5B language model, engineered for lightweight local deployment and an Arch Linux-first developer experience.
All-time downloads
44
Public
Parameters
1.5B
3.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.1 GB · 100%
At a glance
- Model type
- qwen2
- Access
- Public
- Created
- Jul 7, 2026
- Updated
- Jul 7, 2026
- SHA
- ebe6b398
Model card
From the Hugging Face model README
🌙 Luna-2.5B
Luna-2.5B is a customized and optimized derivative of the open-source Qwen2.5-1.5B language model, engineered for lightweight local deployment and an Arch Linux-first developer experience.
Rather than changing the underlying model architecture, Luna-2.5B focuses on:
- ⚡ Hardware optimization for lightweight systems
- 🐧 Arch Linux developer workflow
- 💻 Programming and terminal assistance
- 🚀 AI-powered productivity
- 🌙 Antigravity reasoning framework
- 📦 Native Ollama integration
Luna-2.5B is designed to feel lightweight, structured, and developer-friendly while remaining fully compatible with the original Qwen architecture.
🐧 Arch Linux Customization
Luna-2.5B has been customized specifically for Linux users and developers.
Optimizations
- Arch Linux–first responses
- Native pacman and AUR knowledge
- Hyprland workflow awareness
- Bash & Zsh optimized
- System administration assistance
- Python development
- FastAPI
- Docker
- Git
- Node.js
- Developer tooling
The model prioritizes Linux-native solutions whenever appropriate while remaining cross-platform.
🌙 Antigravity Philosophy
Antigravity is not about fictional propulsion.
It represents reducing complexity.
Every response should:
• Reduce cognitive load
• Remove unnecessary complexity
• Keep explanations structured
• Make difficult concepts feel effortless
Knowledge should feel weightless.
⚙️ Technical Base
Base Model
Qwen2.5-1.5B-Instruct
Original Authors
Qwen Team
Derivative
Luna-2.5B
Customization & Optimization
Arunachalam
Architecture
Transformer Decoder LLM
Deployment
Ollama
License
Apache 2.0 (See original model license)
🚀 Install
ollama create luna-2.5b -f Modelfile
ollama run luna-2.5b
🐧 Recommended Environment
Operating System
Arch Linux
Shell
Zsh
Terminal
Kitty
Desktop
Hyprland
Hardware
Intel Core i3+ 8 GB RAM minimum
❤️ Credits
Original Model
Qwen Team
Luna-2.5B Customization
Arunachalam
Luna AI Ecosystem
Created and maintained by Arunachalam.
- Type
- qwen2
- Created
- Jul 7, 2026
- Updated
- Jul 7, 2026