# Inception: Mercury: AI model fact sheet

- **Provider:** inception
- **Released:** 2025-06-26
- **Context window:** 32,000 tokens
- **Parameters:** Not disclosed
- **API pricing:** $10.00 / 1M input, $10.00 / 1M output
- **OpenRouter ID:** inception/mercury
- **Capabilities:** conversation, reasoning, code-generation, analysis

Mercury is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like GPT-4.1 Nano and Claude 3.5 Haiku while matching their performance. Mercury's speed enables developers to provide responsive user experiences, including with voice agents, search interfaces, and chatbots.

## Benchmarks

| Benchmark | Score |
| --- | --- |
| Speed | 5-10x faster than GPT-4.1 Nano |
| Performance | Matches GPT-4.1 Nano and Claude 3.5 Haiku |
| HumanEval | 90.0% |

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/mercury