|
majipa007@github ----------------------------. . OS: ............. Linux, Windows 11, Android . Uptime: ............................ 4 years . Host: ............................ Ascentium . Kernel: ..................... AI/ML Engineer . Location: ................... Chennai, India . Shell: ................................. zsh . Editor: ..................... Neovim, VSCode . . Lang.Programming: ............... Python, Go . Lang.Markup: .............. HTML, YAML, JSON . Lang.Human: ......... English, Nepali, Hindi . - Currently ---------------------------------. . Focus: .................... LLM quantization . inference infrastructure . Goal: ........... making large models run on . machines that shouldn't fit them |
featherquant  |
Out-of-core GGUF quantization under a hard memory budget. Quantize models larger than your RAM, with crash-safe resume and a live TUI. |
inference-gateway-go  |
Production-style Go gateway for ML/LLM inference: admission control, request batching, scheduling, deadlines, observability. |
LLM-Quantization-Lab  |
GGUF quantization experiments with automated benchmarks, reproducible scripts, and Hugging Face model releases. |
Quantization-YOLOv8  |
A guide to quantizing YOLOv8 object detection models with ONNX. |
