Together AI's Rubin kernel blog builds from B200 GEMMs, stacking new ISA features
togethercompute · x · 2026-09-11
Third post in the thread: the blog starts from the original B200 GEMMs and progressively adds Rubin's new features — wider MMA steps, more tensor/shared memory, b-side collectors, early A release, and more. The kernels are still early, with LUT GEMMs, hardware-native megakernels, and new engine optimizations coming next.
Related event: ThunderKittens Lands on Vera Rubin: NVFP4 GEMM Tops 22 PFLOPS(7 posts)→
More from Infra
- Anthropic's $430B compute contracts are noncancelable, topping China's military budget — xiaosun86 · 2026-10-27
- AI Infra Meme: While Countries Need TW-Scale Power for Datacenters, People Argue About Lightbulbs — mgostIH · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Cloud release velocity scores: AWS 95, Azure 90, Google Cloud 82 — DavidLinthicum · 2026-10-27
- Selling a 5090 to scale local inference: one business weighs 5 hardware paths — Anarchaotic · 2026-10-27
- AWS tutorial: fleet-wide edge CV model lifecycle with SageMaker SDK v3, ~$0.07 per run — AWS ML Blog · 2026-10-27