Skip to main content

Welcome to the Token Plan!

MiniMax is one of the few AI labs that develops frontier models across the full spectrum of modalities: text, speech, video, image, and music. With the Token Plan, you receive a generous usage quota which can be used across MiniMax models of all modalities. The Token Plan was made to satisfy all kinds of builders, from short-film creators to voice agent architects to software engineers.

Core Advantages

  1. Full-Spectrum Multimodal Access — Unified access to MiniMax’s entire model lineup (text, speech, video, image, and music) under a single quota.
  2. Powered by the Latest M2.7 Model — All packages include the latest MiniMax M2.7 model, with M2.7-highspeed availability based on resource load. High-Speed subscriptions offer dedicated M2.7-highspeed support for even faster inference.
  3. Extremely Cost-Effective — A fixed-fee subscription grants you a substantial number of requests usable across all supported models, so you can build freely without unpredictable bills.

Usage Quota

The usage quota provisioned by the Token Plan is measured in requests. One request is roughly equal to one call to M2.7. We offer a variety of plans (Plus, Max) to meet diverse needs. Additionally, we offer High-Speed plans (Plus-Highspeed, Max-Highspeed, Ultra-Highspeed) with dedicated support for the MiniMax-M2.7-highspeed model, which consumes 2 requests per call. Here is the request usage of other models:
ModelInference Spend
Text to Speech HD600 requests per 1000 characters
Text to Speech Turbo300 requests per 1000 characters
Async Text to Speech HD450 requests per 1000 characters
Async Text to Speech Turbo225 requests per 1000 characters
MiniMax-Hailuo-2.3-Fast3000 requests per 768P, 6s video
MiniMax-Hailuo-2.3
MiniMax-Hailuo-02
4500 requests per 768P, 6s video
MiniMax-Hailuo-021500 requests per 512P, 6s video
MiniMax-Hailuo-022250 requests per 512P, 10s video
Music-2.5+3000 requests per 5-min music
Music-2.53000 requests per 5-min music
Music-2.0450 requests per 5-min music
Lyrics75 requests per song
image-01 / image-01-live75 requests per image
When a generation is made from any of these models, the equivalent number of requests will be deducted from your plan’s 5-hour quota.

Getting Started

Only 2 steps to quickly activate your Token Plan service
1

Subscribe

Visit the Token Plan Subscription page, choose the plan that suits your best
2

Get Token Plan API Key

Access Account/Subscription to get Token Plan API Key
Step 1: Subscribe Visit the Token Plan Subscription page, choose the plan that best suits your needs and complete the subscription process. subscribe Step 2: Get Token Plan API Key Once your subscription is successful, navigate to the Account/Token Plan page. Here you can view your active plan details and get your Token Plan API Key. Api key
Important Notes
  • This API Key is exclusive to the Token Plan and is not interchangeable with the API Keys for pay-as-you-go text models.
  • This API Key is only valid during the active period of your Token Plan subscription.
  • Please protect your API Key to prevent any loss of resources.

Use in AI Agents and Coding Tools

After obtaining your API Key, you need to configure it in your preferred AI coding tool. For detailed setup instructions, please refer to our official documentation: Please select the configuration guide corresponding to the tool you use (e.g., OpenCode, OpenClaw, etc.).

After Reaching the Usage Limit

When you reach the request limit within a 5-hour window, you have the following options:
  1. Switch to Pay-As-You-Go: If you wish to continue without rate limits, you can replace your Token Plan API Key with your pay-as-you-go API key. This will switch the tool to a pay-as-you-go model based on actual token usage, which will consume your API account balance.
  2. Wait for the Reset: The Token Plan’s usage limit is calculated on a rolling 5-hour window. You can simply pause usage and wait for your quota to be restored.