AI·DominoTree is a training-free best-first tree draft builder for speculative decoding that scores each candidate node by re-applying Domino's GRU-based causal correction along its specific root-to-node path. It restricts per-node correction to top-M candidates for efficiency and delivers up to 6.6x speedup with the highest mean accept length of any tested method.
Search
What are you looking for?
2 results for “llm-inference”
AI·User compares 5090 at 600W full compute and prompt processing vs shunt-modded 6000 PRO MaxQ water-cooled at 300-600W.