Grok 4

xai · Ranked across 5 benchmarks · best rank #4

Benchmark scores

BenchmarkCategoryRankScoreCaptured
Aider Polyglot code #4 79.6% 2025-07-11
Terminal-Bench 2.0 agents #11 61.8% 2026-04-01
AA Coding Index code #17 42.2% 2026-06-15
Chatbot Arena chat #19 1461 2026-06-10
OpenRouter · Weekly Usage usage #24 #56 2026-06-09