Run two system prompts or two Claude models on the same inputs and compare answers, speed and cost.
Runs in your browser with your own Anthropic API key. No server, no account.
Two variants, each with its own model and system prompt
Any number of test inputs, both variants run in parallel on each
Time to first token, total time, tokens and estimated cost per answer
Summary table and CSV export of every run
This page has no server. It calls the Anthropic API straight from your browser with your own key, which is sent only to api.anthropic.com. Each run shows the tokens used and an estimated cost.