DSpark and DFlash2 draft models for faster inference with speculative decoding
Yosef Worku Alemneh
rasyosef
AI & ML interests
Pretraining, Supervised Fine Tuning, Direct Preference Optimization, Retrieval Augmented Generation (RAG), Function Calling
Recent Activity
new activity about 18 hours ago
IFM/K2-Horizon-7B:⚡ Up to 5.1x faster inference with DSpark speculative decoding (71 --> 364 tokens/sec) new activity about 23 hours ago
IFM/K2-Horizon-7B-Uno:⚡ Up to 5.1x speedup with DSpark speculative decoding (71 --> 364 tokens/sec) updated a model 1 day ago
rasyosef/K2-Horizon-7B-DSpark