MiniMax API Platform

What it does

MiniMax's developer platform exposes models through an API. You send text prompts, images, video, audio or reference material and receive generated text, video, synthesized or transcribed speech, and images. The pricing page lists language models (MiniMax-M3.1-Flash, MiniMax-M3), the H3 and Hailuo video model families, text-to-speech and speech-to-text, voice design and cloning, an image model (image-01), and an API-vlm tool.

Developers call it from code through the Anthropic SDK (recommended), the OpenAI SDK, or native endpoints, setting a base URL and API key as environment variables. Billing is per token, character, second of video, image, voice or request, so one account covers several modalities. Audio and video packages are available as alternatives, and the M3 language model offers a Priority service tier at 1.5x standard rate.

It suits developers and teams building apps that mix language, speech and video generation. You register on the platform, create a key and add balance, or buy an M Plan or Credits. Limits: video packages do not yet cover H3, paid music generation APIs close to new users from August 20, 2026, and there is no consumer interface here.

What it can do

• Text generation with MiniMax-M3 and M3.1-Flash, including prompt caching and a Priority tier
• Video generation (H3 and Hailuo families) from text, images, video or audio references
• Text-to-speech, plus voice design and rapid voice cloning
• Speech-to-text transcription with speaker diarization and srt/vtt export
• Image generation and an API-vlm MCP tool

Works with

• Anthropic SDK
• OpenAI SDK
• AI SDK

Example workflows

Add a language model to an existing app

Input: an app already using the OpenAI SDK. Steps: create a pay-as-you-go key, set OPENAI_BASE_URL to https://api.minimax.io/v1 and OPENAI_API_KEY, then send a chat request. Output: generated text billed per million input and output tokens.

Produce narrated clips

Input: a script and reference images. Steps: synthesize speech with a TTS model, generate video from the images with an H3 model, then combine the outputs. Output: a speech file and a short video, billed per character and per second.

Before you choose

• Video packages cover Hailuo models only; H3 requires pay-as-you-go or a custom plan
• Paid Music and Lyrics APIs close to new users from August 20, 2026, and free music APIs are discontinued
• Pay-as-you-go keys and M Plan Subscription Keys are different, and the Subscription Key only works once seats or Credits are available

Listed under

• Creative
• Music and audio
• Video generation
• AI models
• API
• API for developers
• Coding agent
• Developers
• LLM

Get started

Register, create a key, then top up balance or buy an M Plan

1. Register or log in to the MiniMax API Platform
2. Create a secret key under API Keys (or find your Subscription Key under Billing > M Plan)
3. Top up balance or buy M Plan seats or Credits, then set the key as an environment variable

Pay-as-you-go draws from account balance. Units vary: tokens, characters, seconds, images, voices or requests. Priority tier for MiniMax-M3 costs 1.5x standard. See the pricing overview for full rates.

Source: platform.minimax.io/docs/pricing/overview, 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.

• Prerequisites - MiniMax API Docs ↗platform.minimax.io
• Pricing Overview - MiniMax API Docs ↗platform.minimax.io

Company website

platform.minimax.io: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

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