Run prompts against GPT-4, Claude, and Gemini simultaneously. Version them with git-like diffs. A/B test variations. All local-first — your API keys never leave your browser.
No account required. Bring your own API key.
Send the same prompt to GPT-4, Claude, and Gemini simultaneously. Compare outputs side-by-side in real time with response times and token counts.
Every edit creates a version. Compare changes with git-like diffs. Track how your prompts evolve and roll back to any previous version.
Organize prompts with folders and tags. Search across your entire collection. Export as JSON, Markdown, or plain text.
Create variations of your prompts and test them against each other. See which phrasing produces better results across all your models.
All data stored in your browser via IndexedDB. API keys encrypted at rest. No server, no tracking, no account needed.
Export test results and prompts as JSON, Markdown, or plain text. Integrate with your existing workflow.
Lifetime access. No subscription. Free updates.
Enter your OpenAI, Anthropic, and/or Google API keys. They're encrypted and stored locally.
Type or paste your prompt. Hit Run to test it across all enabled models simultaneously.
Compare outputs side-by-side. Version your prompts. A/B test variations to find the best results.