Speculative Decoding and Efficient LLM Inference with Chris Lott - #717

by The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)