The AI model catalog. One API, every provider.
Find models by what they make, who builds them, or what you need them to do.
Compare models side by side. See how models differ before you choose one.
Compare modelsRecipes that skip the wiring.
Ready-made workflows that combine models and steps into one production flow. Fork a recipe, change the inputs and run it.
The models keep changing. Your workflow doesn’t have to.
New models appear constantly. each::labs gives you one place to explore them, compare them and build with them.
Discover
Find the model that fits the job.
Compare
See price, speed and capabilities side by side.
Build
Test models before committing to an integration.
Orchestrate
Combine models into repeatable workflows.
Explore the catalog
The short answers. For anything specific to your setup, talk to our team.
Talk to an engineerPlan for your expected volume.Models are organized by what they do, the provider behind them and the type of workflow they support. You can search, filter and compare models based on your production needs.
No. Browsing the catalog and reading model pages is open to anyone. You only need an account to run a model or a workflow in your own project.
Filter by task or provider, or search for what you want to make. Each card shows what the model is best for, and its page has examples, inputs and pricing.
Yes. Every model page shows its price and the unit it is billed in, per image, per second or per token.
Open Compare to run the same input on several models and see their outputs, speed and cost side by side. Comparing runs real models, so it needs a free account.
The catalog updates continuously. New models are added as providers release them, and deprecated models are flagged.
