view article Article Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding nvidia • Mar 19 • 50
view article Article Nemotron 3 Nano 4B: A Compact Hybrid Model for Efficient Local AI nvidia • Mar 17 • 71
LLM Pruning and Distillation in Practice: The Minitron Approach Paper • 2408.11796 • Published Aug 21, 2024 • 62