NINA 8B — NinaCatCoin AI Model (VPS Edition)
Fine-tuned LLaMA 3.1 8B model for NinaCatCoin cryptocurrency AI assistant. Optimized for VPS deployment with CPU inference (24GB RAM).
Model Details
| Property | Value |
|---|---|
| Base Model | Meta LLaMA 3.1 8B Instruct |
| Method | QLoRA (r=16, alpha=32) |
| Dataset | 3,980 multilingual pairs (EN/ES/FR/DE/PT) |
| Epochs | 3 |
| Format | GGUF Q4_K_M |
| Size | 4.6 GB |
| SHA-256 | CF93DE6C27406C0C03958700498E6888F12C9BE94208148F4BE7587EE2CF542E |
Download
`ash
Direct download
wget https://huggingface.co/ninacatcoin/nina-model-8b/resolve/main/nina-8b-nina-ft-Q4_K_M.gguf `
Usage with llama.cpp
ash ./llama-server -m nina-8b-nina-ft-Q4_K_M.gguf -c 4096 --port 19021
What is NinaCatCoin?
NinaCatCoin is a Monero-based cryptocurrency with 34 embedded AI modules. NINA is the AI assistant that runs on every node — no cloud, fully decentralized.
Related
- 3B Model (lightweight): ninacatcoin/nina-model
- GitHub: NinaCatCoin
- Downloads last month
- 24
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for ninacatcoin/nina-model-8b
Base model
meta-llama/Llama-3.1-8B
Finetuned
meta-llama/Llama-3.1-8B-Instruct