Open-source memory server that lets AI coding agents like Claude Code and Cursor recall past sessions, with local-first search.

MiniMax API Platform
Developer platform to call MiniMax M3 language, H3 video, speech and image models by API, pay-as-you-go or with a Token Plan.
What is MiniMax API Platform?
The MiniMax API Platform is where developers call MiniMax models over an API. The language models are MiniMax-M3 and MiniMax-M3.1-Flash, multimodal coding models with a 1M-token context window, while MiniMax H3 produces video from text, images, first and last frames, or reference material at up to 2K resolution. The audio section lists speech-2.8-hd and speech-2.8-turbo for text-to-speech in 40 languages, speech-to-text with speaker diarization, plus voice design and rapid voice cloning, and image-01 handles image generation. Billing is published per use: M3.1-Flash costs $0.15 per million input tokens and $0.60 per million output tokens, and speech-2.8-turbo costs $60 per million characters. For predictable spend, the Token Plan subscription offers Plus at $22 per month, Max at $55 per month and Ultra at $132 per month, with a shared quota for text, image and speech that works in coding tools such as Claude Code and Cline. Prepaid credits are sold from $5 and stay valid for 365 days.
Use cases
- Call M3 models from an application
- Generate video from text or images
- Convert text to speech via API
- Transcribe audio with speaker diarization
- Power coding agents on a subscription
- Clone or design custom voices
- Generate images by API
Who it's for
Why use MiniMax API Platform
- Pay-as-you-go rates are published per million tokens, per second of video and per character, so costs can be estimated up front.
- A single Token Plan quota covers text, image and speech, which keeps agent spending predictable.
- Subscription keys work in coding tools and any OpenAI-compatible client, so existing workflows carry over.
- A 1M-token context window suits long documents and large codebases.
- Prepaid credits can be bought without a subscription and cover overflow when a plan quota runs out.
MiniMax API Platform features
- MiniMax-M3 and M3.1-Flash with 1M-token context
- Tunable thinking depth on M3.1-Flash
- H3 video generation up to 2K resolution
- Text, image, first-last-frame and reference video modes
- speech-2.8-hd and speech-2.8-turbo text-to-speech
- Speech-to-text with speaker diarization and srt/vtt export
- Voice Design and Rapid Voice Cloning
- Prompt caching and priority service tier
- Token Plan subscription with shared multimodal quota
- Prepaid credits valid for 365 days
- Web search tool for coding scenarios
MiniMax API Platform integrations
How to get started with MiniMax API Platform
- Create a MiniMax API Platform account and open the console.
- Create a pay-as-you-go API key, or buy a Token Plan or credits to get a subscription key.
- Pick a model ID such as MiniMax-M3, MiniMax-H3 or speech-2.8-hd from the models overview.
- Send your first API call, or paste the subscription key into Claude Code, Cline or another compatible tool.
- Watch usage in the console and upgrade, add credits or switch to pay-as-you-go at the limit.
MiniMax API Platform limitations
- Token Plan is meant for individual interactive use; production should use pay-as-you-go
- Token Plan quota runs on 5-hour rolling and weekly windows and does not roll over
- H3, voice design and rapid voice cloning are not covered by Token Plan
- Paid music APIs closed to new users from August 20, 2026
- Legacy models are listed and may be retired
MiniMax API Platform pricing
MiniMax API Platform has no free plan. Plans start at $22 per month, according to the official pricing page on .
MiniMax API Platform plans
Plus
$22/month
Token Plan; $220/year billed annually
- Personal projects and prototyping
- 3-4 agents
- 5-hour rolling and weekly quota windows
Max
$55/month
Token Plan; $550/year billed annually
- Daily coding with agents and multimodal work
- 4-5 agents
- 5-hour rolling and weekly quota windows
Ultra
$132/month
Token Plan; $1,320/year billed annually
- Heavy agent workflows and extended sessions
- 6-7 agents
- 5-hour rolling and weekly quota windows
Plans as listed on the official pricing page on . Prices can change and may vary by country or exclude tax. Check the official pricing page.
MiniMax API Platform alternatives
The main MiniMax API Platform alternatives listed on Debuly are OpenAI API, ElevenLabs, Kimi API Platform, Alibaba Cloud Model Studio, Gemini API and DeepSeek API. The table compares pricing and platforms; order is A–Z.
| Tool | Pricing | Starting price | Platforms |
|---|---|---|---|
| MiniMax API Platform | Paid | $22/mo | API, Web, CLI |
| Alibaba Cloud Model Studio | Paid | $6/mo | Web, API |
| DeepSeek API | Paid | — | API |
| ElevenLabs | Freemium | $6/mo | Web, API |
| Gemini API | Freemium | — | Web, API |
| Kimi API Platform | Paid | $2, usage-based | API, Web |
| OpenAI API | Paid | $0.1, usage-based | API, Web |
Alibaba Cloud Model Studio
Alibaba Cloud platform giving developers API access to Qwen and partner models for text, image, video and speech, billed per token or by subscription.
DeepSeek API
Pay-per-token API for DeepSeek V4 models with a 1M-token context, thinking modes, tool calls and OpenAI or Anthropic formats.

ElevenLabs
AI audio platform for text to speech, voice cloning, dubbing, transcription, music and conversational voice agents, via app or API.
FreemiumVisit ElevenLabs (official site)
Gemini API
Google's developer API for calling Gemini models from your own apps, with a free tier, pay-as-you-go pricing and Google AI Studio.
FreemiumVisit Gemini API (official site)
Kimi API Platform
Pay-as-you-go API for the Kimi K3.1 Pro and K3 models with a 1M-token context, tool calling and built-in web search, memory and code tools.

OpenAI API
Developer API for OpenAI's GPT models, image generation, realtime voice and transcription, priced per million tokens with batch and flex options.
MiniMax API Platform FAQ
What is the MiniMax API Platform?
It is MiniMax's developer platform for calling its language, video, speech and image models by API, with documentation, pricing and a Token Plan subscription for coding agents.
How much does the MiniMax API cost?
Pricing is per use. MiniMax-M3.1-Flash is $0.15 per million input tokens and $0.60 per million output tokens, and speech-2.8-turbo is $60 per million characters. Image generation with image-01 is $0.0035 per image.
How much is the MiniMax Token Plan?
Token Plan is $22/month for Plus, $55/month for Max and $132/month for Ultra. They share one quota across text, image and speech, with 5-hour rolling and weekly windows.
Is there a free plan?
No free plan is listed on the pricing pages. The first 5 input images per video task are free, and audio input to video is free.
Which models can I use?
Language models MiniMax-M3 and M3.1-Flash, H3 video models, speech-2.8 audio models and image-01, plus legacy Hailuo video and earlier speech models.
Can I use MiniMax models in coding tools?
Yes. A Token Plan subscription key can be dropped into tools such as Claude Code, Cline or any OpenAI-compatible client, and MiniMax Code runs directly on the plan.
What are alternatives to the MiniMax API Platform?
Alternatives include the OpenAI API, ElevenLabs, Kimi API Platform, Alibaba Cloud Model Studio and the Gemini API for models via API.
Sources & verification
Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.
Image credits
- Logo: from platform.minimax.io · fetched · Vendor trademark, used only to identify the product; removed on request
- Cover: none collected yet, so a generated tile is shown.
Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.



