Rebellions' 2048 TFLOPS NPU lands major Japanese AI datacenter deal with ai&
DavidBennett__ · x · 2026-09-30
Korean chip startup Rebellions announced its inference-focused NPU Rebel100 will be deployed at scale in Japanese AI datacenter startup ai&'s facilities.
- Rebel100 uses a 4-chiplet design fabbed on Samsung 4nm, delivering 2048 TFLOPS (FP8) per chip with 144GB HBM3e and up to 600W TDP, just manageable with air cooling.
- CBO Marshall Choy stressed a "no proprietary fork" philosophy: the software stack is 100% open source to avoid vendor lock-in.
- The deal reflects a broader trend of heterogeneous compute (NPU/CPU) replacing GPUs for inference workloads at lower power and cost.
More from Infra
- Jensen Huang: AI data centers add 10-20GW a year and roughly a million jobs — victor_explore · 2026-09-30
- First SGLang Summit set for Nov 12-13 in SF, with Intel CEO and Lilian Weng speaking — BanghuaZ · 2026-09-30
- Interview with Richard Ho, leader of OpenAI's in-house chip project — bookwormengr · 2026-09-30
- Tenstorrent opens bio-model training: OpenFold3 on Blackhole Galaxy nears DGX H200 at quarter the cost — MoAlQuraishi · 2026-09-30
- On-device inference is finally practical: NPUs shift AI workloads off the cloud — lmoroney · 2026-09-30
- Cognition runs production workloads on Vera Rubin NVL72 with 4.8x token throughput — nordicinst · 2026-09-30