SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding Paper • 2603.18567 • Published Mar 19 • 1
Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning Paper • 2604.12374 • Published 7 days ago • 35
SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding Paper • 2604.09557 • Published Feb 10 • 10
view article Article Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding Mar 19 • 46