>1B tokens/minute/GPU by combining query planner and inference engine

 (modal.com)

7 points | by charles_irl 13 hours ago ago

2 comments