To Infinity and Beyond: ThunderKittens Now on NVIDIA Vera Rubin NVL72!
Imported from official source
We ported ThunderKittens to NVIDIA's Vera Rubin NVL72 and rebuilt our NVFP4 GEMM around the new hardware, taking it from 42% of roofline to over 22 PFLOPS — competitive with cuBLAS and CuTe DSL. Here is what changed in the ISA and how we used it.
This version
- Version
- 1 of 1
- Recorded
- September 20, 2026 19:52
- Change
- Initial
- Content hash
2f48b2fb2eea4eb387615c97384b2670- All versions
- Revision history