Tyler Storm
480 posts
When one brain isn't enough, switch to Grok 4.20. Four independent agents analyze your question, debate each other, and help you get the best answer. Available now to SuperGrok and Premium+ subscribers globally.
I built my March Madness bracket using Grok 4.20's multi-agent collaboration system, and the process was mind blowing. Grok was able to run a full team of customized agents in realtime to conduct the best analysis possible. Here's how I set it up:
Grok 4.20 beta1 (single agent) debuts #1 on Search Arena, and #4 overall in Text Arena! Highlights: - #1 in Search, scoring 1226, leading GPT-5.2 and Gemini-3 - #4 in Text, scoring 1492 on par with Gemini 3.1 Pro Congrats to the @xAI team and @elonmusk on this impressive milestone!
Since xAI was formed just 30 months ago, the small and talented team has made remarkable progress. The future has never looked more exciting!
📈In October, we opened ForecastBench, our AI forecasting benchmark, to external submissions. Here's how the top two teams approached the benchmark: • @xai: Minimal scaffolding: give Grok 4.20 (Preview) the question, web/X search, Python REPL, average 8 forecasts • @cassi: Multi-stage pipeline: split to sub-questions, retrieval, model ensemble (o3 + GPT-5), crowd adjustment Both are tied at #2 on our leaderboard, behind only superforecasters, and outperforming our baseline LLM runs.
It was great hosting an amazing group of hardcore engineers for the last 24 hours. Here are the highlights of the top projects. 🚀🧵
