Fusion
Fusion is an AI model provided by OpenRouter that processes prompts by utilizing a panel of expert models. It performs parallel analysis combined with integrated web search and web fetch capabilities to generate responses.
Fusion specs and pricing
- Provider: OpenRouter
- Released: June 13, 2026
- Context window: 1M tokens
- Input: text
- Output: text
- Price: — input / — output per 1M tokens
What Fusion is good at
- Researching complex topics using live web data
- Synthesizing multiple expert perspectives on a single prompt
- Gathering information from various online sources simultaneously
To-dos to try with Fusion
- Search the web for the latest developments in renewable energy and summarize the key findings from multiple perspectives.
- Fetch the content of three different news articles on a specific topic and create a balanced summary of the events.
- Analyze a complex policy question by aggregating insights from multiple expert viewpoints found online.
FAQ
- How does Fusion generate its responses? Fusion functions by running a panel of expert models in parallel that analyze your prompt while simultaneously utilizing web search and web fetch tools.
- What types of input does Fusion accept? Fusion is designed to process text-based inputs.
- How much does Fusion cost? Fusion 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 Fusion? Fusion can read up to 1M tokens at once.
- Can I use Fusion without coding? Yes. In todo.is you pick Fusion 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.