Joylo uses a combination of frontier AI models to power your app generation experience. The models are selected automatically based on the task — you don't need to configure anything.
Which models are used?
Joylo currently uses models from Anthropic (Claude) and OpenAI (GPT). Each model is chosen based on the specific task it performs best:
- Planning and architecture — frontier reasoning models for high-quality structural thinking
- Code generation — models optimised for accurate, idiomatic code output
- Design and UI — specialised image and layout generation models
- Confidence Score review — security-focused analysis models that check generated code for vulnerabilities and best-practice violations
Can I choose which model to use?
Not currently. Joylo automatically routes each step to the best available model. This ensures consistently high-quality outputs without requiring you to understand model tradeoffs.
Will the models change over time?
Yes. As new, more capable models are released, Joylo upgrades its routing automatically. You benefit from model improvements without any action on your part.