Unconventional AI claims 1,000x power cut for inference

unconventional ai, a startup led by former databricks ai chief naveen rao, has introduced un0, an image generation model designed to demonstrate the capabilities of its new oscillator-based computing architecture. currently running via software simulation, the model matches the performance of state-of-the-art diffusion models like stable diffusion but aims to eventually reduce inference power consumption by a factor of 1,000. the company plans to publish schematics for its physical chips soon, aiming to build a full inference stack that addresses the impending energy bottlenecks facing the ai industry.