Unconventional AI shows analog oscillator chip, claims 1000x power efficiency over Nvidia in 2 years

PTrubey · x · 2026-09-16

• At the All-In Summit, Unconventional AI CEO Naveen Rao unveiled its first fabricated proof-of-concept chip, which stores neuron weights not in SRAM/HBM but in an analog dynamic medium of coupled oscillators — compute and memory share the same substrate, eliminating the Von Neumann bottleneck.
• The startup is one of four (with Ineffable Labs, ami, Thinking Machines) that recently raised over $4.5B combined to replace conventional silicon, backprop, and transformers, all attacking AI's power inefficiency.
• The company promises 1000x power efficiency over Nvidia within two years, but disclosed no real benchmarks beyond energy per generated image.

Original post →

More from Infra

• MLPerf Inference v6.1 results imminent: 30 submitters, new accelerators and platforms — TheKanter · 2026-09-16
• Lambda runs 19 nodes in a 16-node power budget with NVIDIA DSX, +24% throughput — TheZachMueller · 2026-09-16
• NVIDIA unveils DSX AI Factory platform to maximize output per megawatt — nvidia · 2026-09-16
• Claim: run a 125B MoE at 22 tok/s on $320 of used GPUs with llama.cpp — cephaloform · 2026-09-16
• Astera Labs' Leo X-Series memory controllers boost long-context inference TTFT by up to 62% — BenBajarin · 2026-09-16
• Dev measures under 4.5GB VRAM peak throughout full local generation — cocktailpeanut · 2026-09-16

Infra channel →