Jason KnightinOctoAIAuto-scheduling now available in TVMSupporting broader model and hardware coverage, and even easier custom op support.Mar 4, 2021Mar 4, 2021
Jason KnightinOctoAICompiling classical ML for (up to 30x) performance gains and hardware portabilityOctoML and Microsoft make classical machine learning even faster.Feb 25, 20211Feb 25, 20211
Jason KnightinOctoAIIn the cloud — Sparsity on GPUs provides 5X speedupAuthor: Tristan Konolige and Sayce FalkJan 15, 2021Jan 15, 2021
Jason KnightinOctoAIUsing Sparsity in Apache TVM to halve your cloud bill for NLPBy Joshua Fromm, Bing Xu, Morgan Funtowicz and Jason KnightJul 17, 2020Jul 17, 2020
Jason KnightinOctoAITinyML: TVM taming the final (ML) frontierMachine learning on embedded platformsJun 4, 2020Jun 4, 2020
Jason KnightinOctoAIWebGPU powered machine learning in the browser with Apache TVMEasy ML deployment in the browser with near-native GPU performance and WebAssemblyMay 14, 2020May 14, 2020
Jason KnightinOctoAIOctoML ≔ Easier machine learningMachine learning and deep learning (ML/DL) are making large impacts across the computing field and the horizon is bright with the rapidly…Mar 24, 2020Mar 24, 2020