Local AI 202: Get Massive Speed-ups with Speculative Decoding

Video by Hugging Face via YouTube
Local AI 202: Get Massive Speed-ups with Speculative Decoding

Learn about speculative decoding techniques like EAGLE, DFlash, MTP with llama.cpp, transformers and speculators library in this livestream.
Schedule:
18.10: Introduction to Speculative Decoding (Merve Noyan, Hugging Face)
18.30: Getting Started with Speculators (Eldar Kurtić, RedHat)
19.00: Speed-up Gemma-4 with MTP Drafters (Maarten Grootendorst, Google DeepMind)

Source