Inception
First-party API for Mercury diffusion LLMs
Last reviewed Oct 20, 2026
Inception develops the Mercury family of diffusion large language models (dLLMs), which generate text by refining a whole draft in parallel rather than one token at a time, and sells first-party access to them through the OpenAI-compatible Inception API.
We're actively tracking prices for Inception. Check back soon, or browse other providers with current pricing.
Pros & Cons
Advantages
- First-party access to the Mercury diffusion models
- OpenAI-compatible API
- Cached-input rates published per model
- Also available through OpenRouter, AWS Bedrock and Azure Foundry
Limitations
- Only Inception's own models are offered
- Older Mercury models are closed to new users
Key Features
Mercury chat models
Chat Completions endpoint for the Mercury models, with tool calling, structured outputs and selectable reasoning effort
Code completion and next-edit
Fill-in-the-middle and edit endpoints served by the Mercury Edit code model for IDE autocomplete
Mercury Decide
A decisions endpoint for classification and scoring over shared context, returning calibrated probabilities
Prompt caching
Cached input tokens are billed at a reduced rate
Pricing Options
| Option | Details |
|---|---|
| Per token | Input, cached input and output tokens billed per 1M |
| Enterprise | Dedicated capacity, private networking and custom terms through sales |
Availability & Support
Support
Documentation, Discord community and enterprise sales
Company
Published by Inception to identify the company behind this listing.
- Legal entity
- Inception AI, Inc.
- Headquarters
- 🇺🇸 Redwood City, United States
Getting Started
- 1
Create an account
Sign up on the Inception Platform
- 2
Create an API key
Generate a key in the platform dashboard
- 3
Call the API
Send requests to api.inceptionlabs.ai with an OpenAI-compatible client or the Inception SDKs
Frequently Asked Questions
Which models does Inception serve?
Check the pricing table above for the models Inception currently serves and their per-token rates.
How do I get started with Inception?
Create an account, Create an API key, Call the API
What are Inception's main advantages?
Inception's main advantages include: First-party access to the Mercury diffusion models, OpenAI-compatible API, Cached-input rates published per model, Also available through OpenRouter, AWS Bedrock and Azure Foundry.
What are Inception's limitations?
Inception's main limitations include: Only Inception's own models are offered, Older Mercury models are closed to new users.