YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
NGARi Qwen3.8-27B Sovereign Edge Model
Sovereign AI: All inference runs on user-owned edge hardware (NGARi Orin 64GB). Zero cloud dependency. Verified air-gap. Apache 2.0 kernel. "AI You Own. Completely."
This repository contains the model card and configuration for running Qwen3.5-27B-Instruct at 4-bit Q4_K_M quantization on NGARi Orin 64GB edge hardware.
Key Features: • Runs on NGARi Orin 64GB with 4-bit AWQ quantization (~16GB VRAM) • Zero cloud dependency — all inference on owned hardware • Verified air-gap: zero bytes transmitted (monitored via /proc/net/dev) • Apache 2.0 kernel with open-source edge deployment • Per-token pricing: $0.50 / 1M input, $1.00 / 1M output
Usage via Ollama:
ollama run qwen3.8:27b "your prompt here"
Sovereignty Principles: • AI You Own. Completely. • Zero cloud dependency for inference • Verified air-gap compliance • Edge-grade hardware sovereignty • Apache 2.0 licensed kernel with open-source deployment