[Inception Releases Diffusion Large Model Mercury 3, Supporting Million-Token Context] According to an official announcement from Inception, the company has released the diffusion-based large language model Mercury 3, which supports a context of 1 million tokens. The official statement claims that its generation speed on NVIDIA GPUs reaches 1,068 tokens per second, with performance on multiple benchmarks surpassing GPT-6 Luna and Claude Haiku 5.5. Standard pricing is set at $0.20 per million input tokens and $0.70 per output token, with a 50% discount during the initial launch period. The model is now available via Inception API, OpenRouter, Baseten, and Vercel, and is compatible with the OpenAI API.