Kimi K3 was partly trained on Blackwell GPUs, per The Information
teortaxesTex · x · 2026-07-28
A quote tweet questions how useful Blackwell GPUs really are for training V4, arguing they are cheap on hardware less than four years old and that 1e25 is not a huge amount of compute. The cited claim says Kimi K3 was partly trained on Blackwell GPUs alongside Qwen3.8 and DeepSeek, according to The Information.
Related event: Moonshot Reportedly Used Nvidia Blackwell for Kimi K3(2 posts)→
More from Infra
- Meta's commitments jumped to $349B in a quarter as server memory prices tripled ahead of Oct 21 earnings — tengyanAI · 2026-10-16
- Meta releases Muse Spark Open, trained with vLLM-powered RL inference from day one — colinmcnamara · 2026-10-16
- Meta's open Muse Spark needs 493GB of memory — far beyond any consumer laptop — TejasKumar_ · 2026-10-16
- CoreWeave plunges 20% as credit markets price >60% default odds; is this dot-com all over again? — kimmonismus · 2026-10-16
- MiniCPM-V 4.7 35B-A3B hits 17+ tokens/s on iPad with ~3GB memory — techNmak · 2026-10-16
- China rolls out 'token factories' and a 2026-28 token economy action plan — nordicinst · 2026-10-16