Independent validation is the best kind. @OmerShlomovits got the full @Kimi_Moonshot K3 model running on a single @AMD 8xMI325X node on DigitalOcean, and wrote up every fix it took to get there. 🌩️ Worth a read if you're evaluating inference infra for large MoE models. 👇

In a @CNBC interview today about Q2 earnings, our CEO @Paddix told @KellyCNBC on @CNBCTheExchange that our AI Inference Services ARR grew ~800% year over year. Per today's earnings release, our AI Customer ARR grew 212% year over year to $234 million. 📈 Video

Catch the full interview here. 🔗 do.co/4zd9YTE Link DigitalOcean CEO on earnings: Pursuing the AI infrastructure trade but doing it profitably Paddy Srinivasan, DigitalOcean CEO, joins 'The Exchange' to discuss the company's growth, the business operations and the length…

Q2 is one of my favorite quarters in @digitalocean's 15 year history and certainly in my 11 Qs here. The stat I keep coming back to isn't revenue. It’s that we added over 6000 Inference Engine customers in <90 days, open weight models are driving nearly 75% of token traffic…

Our customers are building AI applications & agents across inference and core cloud services. 🤝 Our Q2 earnings prove that an integrated AI platform makes this much easier: 📈 Overall revenue up 29% YoY 🤝 AI customer ARR up 212% YoY ⚡ 85% of our AI customer ARR now comes…