Mercury 2
by Inception
Mercury 2 is a model from Inception, released on February 24, 2026. It has a 128K-token context window and can generate up to 8K output tokens. API pricing is $0.25 per 1M input tokens and $0.75 per 1M output tokens. It ranks #111 of 220 models by Intelligence Score with a score of 23.5.
Mercury 2 is the fastest reasoning LLM, built on diffusion-based language model (dLLM) architecture. Instead of generating text token-by-token, it refines multiple text blocks simultaneously, achieving over 1,000 tokens per second on Nvidia Blackwell GPUs — 5x faster than leading speed-optimized LLMs. Supports tool usage and JSON output with 128K context window.
Key specs
| Provider | Inception |
|---|---|
| Release date | 2026-02-24 |
| License | proprietary |
| Context window | 128K tokens |
| Max output | 8K tokens |
| Input price | $0.25 / 1M tokens |
| Output price | $0.75 / 1M tokens |
| Intelligence Score | 23.5 |
| Coding index | 12.2 |
Quick answers
What is the context window of Mercury 2?
Mercury 2 has a context window of 128K tokens, with a maximum output of 8K tokens per response.
How much does Mercury 2 cost?
Mercury 2 costs $0.25 per 1M input tokens and $0.75 per 1M output tokens through the API.
Is Mercury 2 open source?
No. Mercury 2 is a proprietary model from Inception (proprietary license).
When was Mercury 2 released?
Inception released Mercury 2 on February 24, 2026.
How does Mercury 2 compare to Claude Opus 5.5 (Reasoning)?
Mercury 2 scores 23.5 on the Intelligence Score versus 60.5 for Claude Opus 5.5 (Reasoning), the current #1. Input pricing is $0.25 vs $4 per 1M tokens.