You May Also Enjoy
ML4LM — Speculative Decoding — from where we left off
less than 1 minute read
Published:
Most blogs stop at the basics and skip the real details. I break down what’s usually missing: batching, accept/reject checks, and fallbacks.
ML4LM — Profiling torch.compile on DenseNet-121 Inference (GTX 1650) [medium]
less than 1 minute read
Published:
Introduction
ML4LM — Guards vs Graph Breaks in PyTorch: What You Need to Know [medium]
less than 1 minute read
Published:
Guards vs Graph Breaks in PyTorch torch.compile
ML4LM — A tiny Triton primer (toy example) [medium]
less than 1 minute read
Published: