tokenstat
tokenstat

Compare

Grok 4.5 vs Grok 4.20 Multi-Agent

List rates and scores as of . Lower $/MTok is cheaper. Higher scores rank better. Input edge: Grok 4.20 Multi-Agent Output edge: Grok 4.20 Multi-Agent

Model A

Grok 4.5

xAI
xai/grok-4-5
reasoningtools
  • List input$2
  • List output$6
  • Context1M
  • Offers35
  • AA intelligence38.8
  • AA coding72.4
  • AA agentic41.2

On plans: Cursor Business, Cursor Pro, Cursor Start, Cursor Ultra

Model B

Grok 4.20 Multi-Agent

xAI
xai/grok-4.20-multi-agent
reasoningtools
  • List input$1.25
  • List output$2.50
  • Context2M
  • Offers12
  • AA intelligenceN/A
  • AA codingN/A
  • AA agenticN/A

On plans: SuperGrok Heavy

Score sources on /benchmarks.