ls /news · home › news
grep -r --recent /news | head -20

News

162492 articles page 350 of 8125 0 sources 30 min sync cycle

// latest articles 162492 indexed

00:00
2026-11-10
abhik.ai
machine-learning · 15m read · neu

One multiply, one RTX 5090, eight kernels

A hand-written CUDA kernel for a 4096-by-4096 linear layer with a batch of 16 vectors ran in 29 microseconds on an NVIDIA GeForce RTX 5090, down from 748 microseconds for the first naive kernel, according to a lab benchm…

← prev page 350 / 8125 next →
LIVE [news] indexed:162492 page:350/8125 en · ua 2026-05-20 · —