Cerebras runs Kimi K3.1 at 2,000+ tokens per second on wafer-scale engine
rohanpaul_ai · x · 2026-11-08
Cerebras released a video showing Kimi K3.1 running at over 2,000 tokens per second on its wafer-scale engine. The full video is available on Cerebras' YouTube channel.
Related event: Cerebras Runs Kimi K3.1 at Over 2,000 Tokens Per Second(2 posts)→
More from Infra
- Vercel lists DeepSeek V4.1 Flash 1017 at 1,000+ tok/s, but provider speed numbers don't add up — gajesh · 2026-11-08
- RL staleness in agentic training: >100-step off-policy regimes favor labs with huge compute — teortaxesTex · 2026-11-08
- Inside China's 'Stargate': an 800MW Inner Mongolia campus built at one-fifth of US cost — dylan522p · 2026-11-08
- Thieves steal $11.4M of Nvidia hardware with a fake food-delivery truck — eyishazyer · 2026-11-08
- Open-source satmux boosts inflight GEO Wi-Fi up to 6.4x with $0.03/h VM relay — steipete · 2026-11-08
- Minos Bittensor Miner Runs Genomic Research on a Mac and Pays in Subnet 107 Tokens — markjeffrey · 2026-11-08