The AI Loadout
←Back to models
Checked Active: 2026-10-20
Inception

Mercury 3

Agents and live apps where waiting on the model is the problem: browser agents, voice and chat support, tutoring, search answers.

Evaluation Analysis: Released Oct 19, 2026. A diffusion LLM: it drafts many tokens in parallel instead of one at a time. Inception reports 1,068 tokens per second on NVIDIA GPUs and says it matches or beats GPT-6 Luna and Claude Haiku 5.5 while running 4 to 8x faster. Reasoning effort instant, low, medium or high. Text only, with tool calling and structured outputs. OpenAI API compatible, on Inception's API, OpenRouter, Baseten and Vercel. Launch price is 50% off; cached input is $0.01 during the sale ($0.02 after). New Inception API keys come with 100 million free tokens.
Context Window1M tokens (131,072 max output)
Input Price / 1M$0.20 ($0.10 at launch)
Output Price / 1M$0.70 ($0.35 at launch)
Release Date2026-10-19

Specs and prices checked 2026-10-20 against the official page.

Used in these setups

Related skills and personas

Agent skills and bot personas that pair well with Mercury 3.

Browse full directories:All Skills (>35)·All Personas (>15)

Want tested skills and personas installed in one command? See the AI Loadout packs, from $9.