Unified Sciences

Making AI radically more efficient — across kernels, models & workloads.

[Dhruv Kumar @ddhruvkr]

Some pretty amazing results by @dem_fier and the team at @unifiedsciences. The same engine found improvements across different problems: cuBLAS, a full model, NeRF rendering, neural search, and matrix inversion - all of which were verified by the system before being accepted.

[Aseem Raj Baranwal @aseemrb]

Great work by @unifiedsciences! The interesting part for me isn't the speedups, but that reproducibility and correctness checks are built into the search itself, not bolted on after, which in my opinion is the right way to do automated optimization. Kudos @dem_fier !

[Unified Sciences @unifiedsciences]

https://t.co/jEJShkObqn