Mercury 3
Mercury 3 is a diffusion-based language model from Inception for text-based tasks. It supports reasoning and tool calling, has a 1 million-token context window, and is suited for agentic workflows and real-time applications.
Mercury 3 specs and pricing
- Provider: Inception
- Released: October 16, 2026
- Context window: 1M tokens
- Max output: 131K tokens
- Input: text
- Output: text
- Price: $0.10 input / $0.35 output per 1M tokens
- Reasoning: yes
- Tool calling: yes
What Mercury 3 is good at
- Reasoning over long text
- Agent workflows that use tools
- Text-based real-time applications
To-dos to try with Mercury 3
- Read this long report and summarize its main points.
- Review these project notes and list the decisions and open questions.
- Use the available tools to complete the steps in this text-based task.
FAQ
- Who makes Mercury 3? Mercury 3 is made by Inception.
- Can Mercury 3 read images? Mercury 3 accepts text input only.
- How much does Mercury 3 cost? Mercury 3 costs $0.10 per million input tokens and $0.35 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 0.33 credits ($1 = 100 credits).
- What is the context window of Mercury 3? Mercury 3 can read up to 1M tokens at once and write up to 131K tokens in one reply.
- Can I use Mercury 3 without coding? Yes. In todo.is you pick Mercury 3 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.