Open-Source AI | Local LLMs & VLMs | Model Quantization
Creator of `quant-kit`: a complete multi-modal GGUF quantization pipeline. I focus on making powerful AI models accessible and efficient for local hardware.
My work spans across Text LLMs, Vision-Language Models (VLMs), Diffusion models, and Whisper ASR—optimizing them without sacrificing intelligence and pushing the efficiency-quality Pareto frontier for edge AI.