SpecBundle Collection A collection of production-grade draft models for speculative decoding β’ 30 items β’ Updated about 1 month ago β’ 20
view article Article Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding nvidia β’ Mar 19 β’ 49
view article Article You could have designed state of the art positional encoding FL33TW00D-HF β’ Nov 25, 2024 β’ 499
Running on CPU Upgrade 14.1k Open LLM Leaderboard π 14.1k Track, rank and evaluate open LLMs and chatbots