https://store-images.s-microsoft.com/image/apps.46574.e2d258b9-0a4e-44aa-93eb-968623037897.fafd4ca0-b5c8-43eb-9059-bb802fd8921f.cec52147-ab74-41f9-b8f1-b2d09b4f1063

mercury-offer

by Inception

Mercury diffusion based LLMs offers 5-10X faster response times with comparable quality & cost

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 Claude Haiku 3.5 and GPT-4o Mini while matching their performance. Mercury's speed means that developers can stay in the flow while coding, enjoying rapid chat-based iteration and responsive code completion suggestions.

At a glance

https://store-images.s-microsoft.com/image/apps.60119.e2d258b9-0a4e-44aa-93eb-968623037897.fafd4ca0-b5c8-43eb-9059-bb802fd8921f.ab211e5f-1d22-4949-9ba2-01c206b1e213
/staticstorage/20251111.3/assets/videoOverlay_62a424ca921ff733.png
https://store-images.s-microsoft.com/image/apps.34093.e2d258b9-0a4e-44aa-93eb-968623037897.fafd4ca0-b5c8-43eb-9059-bb802fd8921f.73e60f45-09e5-473d-aebe-76b29b3b3fd4