Mercury 3

Name: Mercury 3
Author: Inception Labs

Inception Labs 🧠 Language Model

Inception Labs' most capable diffusion LLM: a 1M-token context, 131,072 max output, tool calling, reasoning_effort instant/low/medium/high (medium by default).

Specifications

Input modalitiestext

Output modalitiestext

Capabilitieschat, tools, coder

Relative speed●●●●○ Fast

Context window1,000,000 tokens

Vision inputNo

Tool callingYes

Streaming outputYes

Input price$0.175 / 1M

Output price$0.6125 / 1M

Pricing

$0.175 / 1M input tokens · $0.6125 / 1M output tokens

Platform credit pricing — live rates on the Models page.

Use Mercury 3 on Zubnet →