The Minutes of Dylan Patel on benchmarks https://minutesof.com/dylan-patel/on/benchmarks/ Everything Dylan Patel has said on benchmarks: 3 verbatim quotes between June 2025 and November 2025, each with a timestamp and a link to the recording it… en Sun, 30 Aug 2026 16:22:20 +0000 Patel says neo clouds that ranked highest on ClusterMax benchmarks booked significantly more revenue. https://minutesof.com/q/9a2acec1-fa93-4f91-b7b2-610ae8acb9f1/ https://minutesof.com/q/9a2acec1-fa93-4f91-b7b2-610ae8acb9f1/ “And since then we just released the second version, but in the middle, the clouds that ranked the highest actually ended up booking way, way, way more revenue.” — Clockwork Fri, 21 Nov 2025 17:18:51 +0000 benchmarks Patel reports AI models improved from under 4% to 70% on Frontier Math in six months. https://minutesof.com/q/9274e9a4-6baa-4e2a-b44a-3c2207f42c82/ https://minutesof.com/q/9274e9a4-6baa-4e2a-b44a-3c2207f42c82/ “Like, can't do all the math, all the problems in frontier math. And the models went from sub 4% to 70% in six months.” — Going Direct Sun, 22 Jun 2025 21:30:00 +0000 benchmarks Patel reports AI models improved from below 4% to 70% accuracy on Frontier Math in six months across multiple https://minutesof.com/q/2bb1c7b0-0d8f-45e8-a0a1-db8ef64742e6/ https://minutesof.com/q/2bb1c7b0-0d8f-45e8-a0a1-db8ef64742e6/ “frontier math. And the models went from sub 4% to 70% in six months. Right? And this is from not just OpenAI, but also Anthropic and Google and DeepSeek.” — Going Direct Sun, 22 Jun 2025 21:30:00 +0000 benchmarks