Inception Launches Mercury 3 Diffusion Model with 1M Context and High Speed

Inception Labs has launched Mercury 3 on OpenRouter, a new diffusion language model achieving over 1,000 tokens per second generation speed with a 1M token context window. The company claims it outperforms GPT-6 Luna and Haiku 5.5 on benchmarks like IFBench, GPQA Diamond, and SWE-Bench Atlas, offering a 50% launch discount.