Mercury 2
Mercury 2 is a reasoning-focused language model developed by Inception, released in March 2026. It utilizes a diffusion-based architecture to generate and refine multiple tokens in parallel, which distinguishes it from traditional sequential models.
Mercury 2 specs and pricing
- Provider: Inception
- Released: March 4, 2026
- Context window: 128K tokens
- Max output: 50K tokens
- Input: text
- Output: text
- Price: $0.25 input / $0.75 output per 1M tokens
- Reasoning: yes
- Tool calling: yes
What Mercury 2 is good at
- Complex logical reasoning tasks
- Applications requiring tool integration
- High-speed text generation workflows
To-dos to try with Mercury 2
- Analyze a complex logical puzzle and provide a step-by-step breakdown of the solution.
- Use tool-calling capabilities to query an external database and summarize the findings.
- Draft a technical report that requires multi-step reasoning to ensure internal consistency.
FAQ
- What makes Mercury 2 different from other language models? It is the first reasoning diffusion LLM, meaning it generates and refines multiple tokens in parallel rather than sequentially.
- Does Mercury 2 support tool usage? Yes, the model is built with native tool-calling capabilities.
- How much does Mercury 2 cost? Mercury 2 costs $0.25 per million input tokens and $0.75 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 0.81 credits ($1 = 100 credits).
- What is the context window of Mercury 2? Mercury 2 can read up to 128K tokens at once and write up to 50K tokens in one reply.
- Can I use Mercury 2 without coding? Yes. In todo.is you pick Mercury 2 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.