Skip to content

xAI

Grok 4

At launch, open to SuperGrok and Premium+ subscribers and through the API.

Announced
Access
Public

Benchmark scores

TestScoreBest on the trackerHow it was measured
GPQA DiamondPhD-level science questionsPhD experts: 70%Saturated87%Gemini 4 Argon 96%Independentdefault settings (grok-4-0709)
ARC-AGI-2visual puzzles, model aloneAverage person: 60%Saturated16%GPT-6 Astra 95%IndependentARC Prize config "Grok 4 (Thinking)" (grok-4-0709)
OTIS Mock AIMEcompetition mathsSaturated84%GPT-5.5 100%Independentdefault settings (grok-4-0709)

Compare every model on the same tests →

Also on these charts

Sources