Naveen Rao's new chip makes an AI image on ~500 nanojoules, thousands of times less than a GPU
victor_explore · x · 2026-09-22
Unconventional AI founder Naveen Rao's new chip generates an AI image on roughly 500 nanojoules, where a GPU spends on the order of a millijoule — thousands of times more. Rao went from concept to working silicon in 5 months, and the first number he published was a joule count, not a benchmark. The chip is that cheap because it never moves information around. Expect more startups to start quoting power instead of speed.
More from Infra
- 63% of Americans oppose data centers in their community, spanning both parties, pollster says — AlexTensor · 2026-09-22
- Toby Ord estimates $20M spent on AI's millennium prize result, $200M for solid data — tobyordoxford · 2026-09-22
- What engineers check before adding a new LLM provider — Rama_Surasani_ · 2026-09-22
- Dev reverse-engineers DLSS 5 neural rendering, reimplements it bit-exact in Vulkan at 7.8ms/1080p — bdsqlsz · 2026-09-22
- M5 Ultra Hits 3740 tok/s Prefill on Qwen, Nearly Double Overnight — EAccelerate_42 · 2026-09-22
- Running 2.78T-param Kimi K3 on a single CPU in 8 GB RAM, no framework — tom_doerr · 2026-09-22