Introducing Mercury 2 – Inception
Mercury 2 is introduced as the world's fastest reasoning LLM, utilizing diffusion-based parallel refinement instead of sequential decoding for instant production AI.Summary
Inception has introduced Mercury 2, touted as the world's fastest reasoning language model, designed to make production AI feel instant by overcoming the bottleneck of traditional autoregressive decoding. Mercury 2 employs a diffusion-based approach, generating responses through parallel refinement over a few steps, resulting in over 5x faster generation compared to sequential methods. This architecture allows it to achieve reasoning-grade quality within real-time latency budgets, shifting the quality-speed curve for production deployments. Key specifications include a speed of 1,009 tokens/sec on NVIDIA Blackwell GPUs, competitive pricing, and features like tunable reasoning, 128K context, native tool use, and schema-aligned JSON output. Mercury 2 is positioned to excel in latency-sensitive applications such as coding/editing, agentic loops, real-time voice interaction, and Search/RAG pipelines, with early adopters praising its impact on responsiveness and efficiency. The model is available now and is OpenAI API compatible, allowing for easy integration into existing stacks.
(Source：Inceptionlabs Ai)