Switching models

Every major model lives in the same chat. The model picker in the top left lists them with speed, context size, and price, so you can choose per task instead of per subscription.

Switching mid-conversation

Pick a new model from the dropdown at any point. The previous history is passed along, the new model sees the full context and responds to it. A common pattern: draft with a fast model, then switch to a large reasoning model for the final pass.

Which model for which task

• Everyday questions: a fast, efficient model is enough and costs a fraction of a cent per answer.
• Code, analysis, long documents: a large reasoning model with a big context window.
• Creative work: a high-end model with strong writing quality.

The picker shows price per model. Since billing is per token, an expensive model on a short prompt often costs less than you expect. Every answer shows its exact cost right below it.

Comparing models

Add a second model to the same conversation and both answer every prompt side by side. Useful when you are deciding which model to standardize on, or when you want a second opinion on the same context.

Parameters and presets

The parameters panel lets you set temperature, maximum output length, and a custom system prompt. Save a combination you like as a preset and it becomes a one-click choice in the picker.

Plan availability

The Free plan includes the smaller models. Pro and Ultra unlock every model. If a model needs a higher plan, the picker says so. Details under Plan and subscription.