Benchmark

Grok 4.7 High in Grok CLI: TwentyDollarAI Standard Benchmarking

Time, tokens, cost and weekly usage for one run, plus the output.

Results

Figures from the Grok CLI’s end-of-run summary.

Measured results: metric and value
Time (total runtime)36m 34s
Tokens (total)4.77M
Cost (API-equivalent)$3.2425
Weekly usageAbout 2%

Models tested

  • Grok 4.7 HighPlan: X Premium ($30)

The prompt

Standard TwentyDollarAI Benchmarking

Evidence

The output: a white dog running through a meadow with a toy (about 5 seconds, silent).

My take

  • Excellent visuals: detailed scene, convincing dog, smooth running.
  • Low usage: about 2% of the weekly allowance.
  • Slow: over 36 minutes.