Keys are stored only in your browser. Each key is sent only to its own provider: OpenAI's API or Anthropic's API, never to Litmiq.
Get one at platform.openai.com. Needed for GPT models and the default judge (GPT-5.4 Mini).
Get one at platform.claude.com. Every model and judge call is billed to your own account with that provider at its rates. Save an empty field to remove a saved key.
Litmiq prompt evaluator — compare OpenAI and Anthropic models side by side
Litmiq is a free, no-signup browser app for prompt evaluation. Compare OpenAI and Anthropic models side by side, or bring a key for Google Gemini, xAI Grok, Mistral, DeepSeek, Groq, OpenRouter or Perplexity, score outputs with an LLM judge, and share results as a link. Your API key stays in this browser; Litmiq has no eval backend. Perplexity is forwarded through a no-store proxy.
No API key yet? Watch a recorded 30-second demo run
Try it in 30 seconds, no key needed
It starts on its own when this block is on screen. Watch a real Litmiq run: one support-bot prompt, 8 test cases, 3 models, scored by an LLM judge. These are recorded outputs from Llama 3.3 70B, Gemma 4 26B, Llama 3.1 8B on Cloudflare Workers AI (13 Nov 2026), replayed in your browser, not generated live.