About
We explain AI in plain words, and we show our work
The chatmodel team writes these guides for people who want to use AI for everyday things without learning the jargon. chatmodel is one app with many AI models inside, so we can put them side by side and show you how they really answer.
How we test
- 1. Everyday tasks, not lab tests We test what people actually ask: a condolence card, a confusing letter, a scam email, a trip, dinner, a bedtime story.
- 2. Exactly the same words Every model gets the same prompt, in chatmodel, one try each. We keep what comes back without editing it.
- 3. We show you the answers You see both real answers side by side, so you can judge for yourself, and run the same test in one click.
- 4. Fair verdicts We judge accuracy first, then how well it followed the request and how useful it would be. We say when a model gets something wrong.
- 5. Kept up to date Every comparison shows the date it was tested and what changed. When a new model arrives, we test it again.
Why you can trust it
chatmodel gives you access to models from many companies, including OpenAI, Anthropic, Google and xAI, in one plan. We don't make these models, and none of these companies pays us to rank them. Because every model is available in chatmodel, we have no reason to prefer one: our only aim is to help you pick the right one for the job.
Small tests have limits. The same model can answer a little differently each time, so we treat winners as a guide, show the full answers, and invite you to run the test yourself.
Found a mistake?
Tell us and we'll fix it. Every correction appears in the page's “What changed” list.
All of these models, in one place
Chat, images, video and tools with the best AI models. Switch between them whenever you want.
Start for free