Auto Router
Auto Router is a service from OpenRouter that automatically selects an AI model for your prompt based on community usage patterns. It supports a wide range of input types, including text, images, audio, video, and files, while offering reasoning and tool-calling capabilities.
Auto Router specs and pricing
- Provider: OpenRouter
- Released: November 5, 2026
- Context window: 2M tokens
- Input: text, images, audio, files, video
- Output: text, images
- Price: — input / — output per 1M tokens
- Reasoning: yes
- Tool calling: yes
What Auto Router is good at
- Users who want to automatically match their prompt to an optimal model
- Tasks requiring multimodal input analysis like video or audio processing
- Complex workflows that benefit from automated reasoning and tool usage
To-dos to try with Auto Router
- Upload a video file and ask the model to summarize the key events
- Provide a complex reasoning task and let the router select the most suitable model
- Upload a document file and use tool calling to extract specific data points
FAQ
- What is the primary function of Auto Router? It automatically routes your request to a model selected based on collective community spending and performance data.
- Can I use this model with different types of media? Yes, it accepts text, images, audio, video, and file inputs, and can produce both text and image outputs.
- How much does Auto Router cost? Auto Router costs — per million input tokens and — per million output tokens through OpenRouter. In todo.is a typical agent step costs about — credits ($1 = 100 credits).
- What is the context window of Auto Router? Auto Router can read up to 2M tokens at once.
- Can I use Auto Router without coding? Yes. In todo.is you pick Auto Router as your agent's model and write to-dos in plain words, on the web, the Android app, WhatsApp or Telegram. No API key needed: credits are included, or you can bring your own key.
JavaScript is required to use the todo.is app.