xAI releases Grok 3 and Grok 3 mini in beta, plus DeepSearch, its first agent
AIMade by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made
Released Grok 3 / Grok 3 mini in beta and its first agent, DeepSearch — AIME 2025 93.3%, GPQA 84.6% (measured by xAI), Chatbot Arena Elo 1402
Why it matters
Split out a reasoning mode behind a Think button. The 200,000-GPU cluster in the announcement was mentioned for training the next model; the announcement does not say it was all used to train Grok 3
Source — Primary sources first; secondary reports are labelled “Press”
Officialx.aihttps://x.ai/news/grok-3Added in Week 35, 2026 · See that week
On the map — ⑥ Reasoning Models line
- Previous stopDeepSeek-R1
- This recordxAI releases Grok 3 and Grok 3 mini in beta, plus DeepSearch, its first agent
- Next stopGPT-4.5
In the model family tree — Grok family · 11 versions
- Earlier versionGrok-2 / Grok-2 mini
- This recordGrok 3 / Grok 3 mini
- Later versionGrok 4 / Grok 4 Heavy