Model details
Mercury 3 is Inception’s latest diffusion LLM. It generates more than 1,000 tokens per second and supports a 1M-token context window. Mercury 3 scales to more than 40 times the size of Mercury 1 while maintaining comparable inference speeds.
Compared with Mercury 2.5, Mercury 3 offers:
A 1M-token context window, up from 260K tokens
More than 1,000 tokens per second
Mercury 3 is built for latency-sensitive production workloads, including browser-use agents, interactive education, real-time search agents, and customer support.
