model benchmarking tool

Run any prompt.
Every model.
At once.

Benchmark claude-3.5-sonnet, gpt-4o, and gemini-1.5-pro side by side. Real responses, real latency, real token cost. Your API keys — zero markup.

6
models
$0
platform cost
<1s
overhead
0
data stored

Simple pricing

Start free. Upgrade when you need more.

Free
$0 / forever
For developers exploring and testing models.
  • 10 comparisons / day
  • All 6 models
  • Your own API keys
  • Comparison history
  • Shareable links
  • Export results
Pro popular
$9 / month
For developers who benchmark daily and need history.
  • 500 comparisons / day
  • All 6 models
  • 90-day history
  • Shareable links
  • Export to CSV / JSON
  • Team workspace
Team
$29 / month
For teams evaluating models together.
  • Unlimited comparisons
  • All 6 models
  • 1-year history
  • Shareable links
  • Export to CSV / JSON
  • Team workspace (coming soon)
api keys — saved in your browser only
ANTHROPIC
OPENAI
GOOGLE
models — click to toggle
prompt
// user message 0 chars
results — ranked by speed
⬡
select models · enter a prompt · run comparison