Inception's Mercury 2.5 diffusion model delivers 1,107 tokens per second with a 40% intelligence boost, challenging GPT-5.6 and Gemini at a

Inception's Mercury 2.5 diffusion model delivers 1,107 tokens per second with a 40% intelligence boost, challenging GPT-5.6 and Gemini at a

Mercury 2.5 usa la diffusione per generare testo in parallelo: 1.107 token al secondo, 260K di contesto e prezzi ridotti.