Estimate GPU memory usage for Megatron-based MoE (Mixture of Experts) and dense models. Use when users need to (1)...
find ~/yzlnew/ -name "*.skill"
Guide for using SLIME (LLM post-training framework for RL Scaling). Use when working with SLIME for reinforcement...
Creates professional TikZ flowcharts with a standardized style (Google Material-like colors, node shapes, and layout).
Write, optimize, and debug high-performance AI compute kernels using TileLang (a Python DSL for GPU programming)....