Chat Pipeline
Last updated 8 Nov 2026
How a message finds its model
- What the message needs — reading an image, searching the web, calling a tool, reasoning through a hard problem.
- Which models can do all of it — only the models in the catalog that support everything the message needs, and that your plan and any BYOK keys allow, stay in the running.
- Price and quality — Auto picks among them by price and quality.
Auto is the default, not a requirement: you can pick a model yourself in the model picker.
What this means for you
You do not need to keep up with model names. When providers change their prices or release new models, routing follows the catalog. Every call writes one billing record that names the model that actually answered, so your usage history stays accurate.
See /docs/doc-model-routing-and-auto for the details.
Watch how it works
AI chat that picks the model, searches the web and does deep research
In ToRun chat you just ask. Auto works out what each message needs and picks the best-value model that can do it; attach a photo and that turn moves to a model that can see it. Pick a model yourself and your choice is honoured. The assistant searches the web mid-answer and attaches its sources, uses tools like reminders and notes, runs Deep Research for bigger questions, and answers place questions with a map, nearest first. Your location is used only for that search.