Open App →
Back to News Feed
Tom's Hardware November 21, 2026 By Shane Downing neutral

Homelabber’s second-gen Optane build uses two RTX 3090s to run 2.8-trillion-parameter Kimi AI model at 3.11 tokens per second

Imported source excerpt from Tom's Hardware. This is publisher material, not original BAB News analysis.

AI / LLMNVIDIA / GPUData Center

Imported source excerpt

A Reddit homelabber’s Optane PMem 200 server now runs Moonshot AI’s Kimi K3 at 3.11 tokens/s with two RTX 3090s and a RAM drive.
Read at Tom's Hardware

Read original BAB News analysis

Related Articles

OLED gaming laptop with RTX 5060 and Intel Core Ultra 7 CPU price crashes by $650 in Black Friday offer — Lenovo Legion 5i laptop drops to $1,349 with free games at B&H

Getting a gaming laptop this holiday season doesn't have to mean spending thousands of dollars, as this modestly priced

Tom's Hardware · November 22, 2026

Alienware Aurora PC with RTX 5070, 32GB of RAM drops to $1,899 in Amazon Black Friday offer — Core Ultra 7 equipped PC is one of the best value prebuilt deals we’ve seen so far

Pick up a gaming PC bargain this Black Friday with this RTX 5070-powered Alienware Aurora rig, which also comes with 32G

Tom's Hardware · November 22, 2026

DataBeyond Showcases AI Sorting Solutions for Circular Supply Chains at COP31

ANTALYA, Türkiye, Nov. 20, 2026 /PRNewswire/ -- Hangzhou DataBeyond Technology Co., Ltd. (DataBeyond), a provider of int

PR Newswire · November 22, 2026

[COVER STORY] 'AI spending binge puts Samsung, SK at risk of memory price collapse'

Samsung Electronics and SK hynix could face a sharp memory price decline if hyperscalers scale back the artificial intel

Korea Times · November 22, 2026