Inception

What it does

Inception offers diffusion large language models (dLLMs) accessed through an OpenAI-compatible API. You send text messages, and the models return text completions, with options for tool calling and structured outputs on Mercury 3. The company also offers Mercury Voice, tuned for real-time voice agents, and Mercury Decide, which takes state and typed questions and returns answers with probabilities and extracts spans from input text.

Developers use the API to build chat, coding, and voice applications where response speed matters. Unlike sequential token-by-token generation, the company says its models produce tokens in parallel. Mercury Edit 2 handles fill-in-the-middle autocomplete and next-edit suggestions for code. Official SDKs for Python and TypeScript are available, and the models work with libraries such as LiteLLM and LangChain.

This suits developers and engineering teams building latency-sensitive products, not end users looking for a standalone chat app. Access requires creating an Inception Platform account and an API key, which includes a free token allowance. Mercury Voice is listed as enterprise-only, and Mercury 2 and 2.5 are being retired. Speed and benchmark claims come from the vendor and should be tested against your own workload.

What it can do

• Chat completions via an OpenAI-compatible API endpoint
• Tool calling and structured outputs with Mercury 3
• Mercury Voice model optimized for real-time voice agents
• Mercury Decide endpoint for classification, scoring, and extraction
• Mercury Edit 2 for code fill-in-the-middle and next-edit suggestions
• Adjustable reasoning effort levels on supported models

Works with

• LiveKit
• Pipecat
• Vapi
• Retell
• AISuite
• LiteLLM
• LangChain
• AWS Bedrock
• Azure Foundry

Before you choose

• Mercury Voice is listed for enterprise customers; confirm access before planning a voice product.
• Mercury 2 and Mercury 2.5 are no longer available to new users; migrate to Mercury 3 and check the docs.
• Pricing is per million tokens with separate input, cached input, and output rates; Mercury 3 launch discount is listed as 50% off.

Listed under

• Coding
• Chat
• Code completion
• Developers
• Enterprise
• Fast inference
• LLM API

Get started

Create an Inception Platform account, generate an API key, then send requests to the API.

1. Create an Inception Platform account at platform.inceptionlabs.ai.
2. Create an API key from the API Keys page; new keys include 100 million free tokens.
3. Export the key as the INCEPTION_API_KEY environment variable, then install the Python or TypeScript SDK or call the REST endpoint.

Mercury 3 input is $0.10 and output $0.35 per 1M tokens at the stated 50% launch discount (regularly $0.20 and $0.70). Mercury Voice is $0.20 input and $0.75 output per 1M tokens. Mercury Decide bills input at $0.02 per 1M tokens, output not charged. Plan names Free, Developer, and Enterprise appear without listed numeric amounts. Check the models page for the current rate table.

Source: inceptionlabs.ai/models, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Inception – When Every Millisecond Matters ↗www.inceptionlabs.ai
• Our Models – Inception ↗www.inceptionlabs.ai
• SDKs - Inception Platform ↗docs.inceptionlabs.ai
• Mercury 2 for Intelligent Voice Applications – Inception ↗www.inceptionlabs.ai
• Models - Inception Platform ↗docs.inceptionlabs.ai

Company website

inceptionlabs.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.