Rebellions’ Rebel100 NPU

Designed for servers and pods.

April 1, 2026

Jon Peddie

Rebellions has built a compelling case for inference-focused AI silicon. Its Rebel100 NPU—a quad-chiplet package with 144 GB HBM3E and roughly 2 PFLOPS FP8 compute that targets data center LLM inference with a power-efficiency story that outpaces Nvidia H100 on TPS/W by a vendor-claimed 3.2×. The company wraps that silicon in two deployable infrastructure products, RebelRack and RebelPOD, and backs them with a fully open-source software stack. Independent benchmarks remain thin, but the architecture is credible, and the platform is shipping. Rebellions built the Rebel100 from the ground up for LLM inference, not as a training GPU adapted for inference

...

Enjoy full access with a TechWatch subscription!

TechWatch is the front line of JPR information gathering service, comprising current stories of interest to the graphics industry spanning the core areas of graphics hardware and software, workstations, gaming, and design.

A subscription to TechWatch includes 4 hours of consulting time to be used over the course of the subscription.

Already a subscriber? Login below

Rebellions’ Rebel100 NPU

Enjoy full access with a TechWatch subscription!

This content is restricted

Login