ASAN: A conceptual architecture for a self-creating (autopoietic), energy-efficient, and governable multi-agent AI system.

An open and practical guide to Edge Vision.

Ternary Quantization for LLMs: Implement balanced ternary (T3_K) weights for 2.63-bit quantization—the first working solution for modern large language models.

An open and practical guide to Edge AI Engineering.

An open and practical guide to Edge Audio.

An open and practical guide to Edge Language

The final project for EdgeAI course at NYCU, focusing on accelerating Llama-3.2-3B-Instruct inference on a single NVIDIA T4 GPU.