Tool directory · Coding Agents · Featured in Issue #1

Mercury 3

Last checked Oct 22, 2026

A diffusion-based LLM that Inception says runs at more than 1,000 tokens per second with a 1M-token context window. It's OpenAI API compatible and available on Inception's API, OpenRouter, Baseten and Vercel.

Our Take

Mercury 3 is for browser agents, voice bots and subagents where speed matters more than raw intelligence. Inception claims 1,068 tokens per second on NVIDIA GPUs, a 1M-token context window and quality on par with GPT-6 Luna and Claude Haiku 5.5. All benchmarks are Inception's own. It costs $0.20/$0.70 per million input/output tokens, and is half that during launch. Source