Today's news, MiniMax's latest large model M3 has been fully integrated into Alipay's TokenPay one-stop full-stack solution, enabling enterprises to achieve "bulk purchasing, one-click distribution" of Token seat management. At the same time, users can complete in-app checkout by simply speaking a sentence through Alipay's AI Pay within the Mix Claw service provided by MiniMax Agent.

MiniMax is one of the first global large model manufacturers to integrate Token Pay payment capabilities into their core subscription system, marking the large-scale application of the first domestic MaaS (Model as a Service) payment solution launched by Alipay.

It is reported that MiniMax will also be fully integrated with Alipay's full-stack AI payment products in the near future, upgrading service experiences for developers and individual users, accelerating commercialization, and enabling developers to "use natural language to call payment capabilities," making it easier for users to command Agents to perform tasks "more obedient and more reliable."

Enterprise-level Token Seat Management

Bulk Purchase, One-Click Distribution

As large models become enterprise productivity tools, more and more companies are purchasing Token services in bulk. However, in actual operations, enterprises face multiple management challenges: how many seats to buy? How to choose the package for each seat? How to efficiently distribute them internally? How to handle reimbursement?

image.png

To address these pain points, Token Pay has upgraded its enterprise seat management capabilities, offering features such as one-click subscription for enterprise groups, seat expansion or reduction, etc., in complex scenarios like group management, token allocation, and group scaling. After integrating with MiniMax, this is equivalent to establishing a complete payment SaaS system, significantly enhancing the purchasing experience for enterprise customers.

Deep Integration with Large Models

M3 Model Demonstrates Strong Security Capabilities in Payment Scenarios

This collaboration not only achieved product-level integration but also completed the optimization of "large model intent understanding" and "financial-grade payment execution" at the underlying level.

On June 1st, MiniMax officially released its new general-purpose model, MiniMax M3, which adopts a new self-developed sparse attention architecture called MiniMax Sparse Attention (MSA). M3 is the first large model in China to simultaneously possess three core capabilities: cutting-edge Coding ability, 1M ultra-long context, and native multimodal support. It is also the only open-source option globally with a complete set of capabilities.

During the adaptation testing of the M3 model, Alipay found that M3 performed exceptionally well in intent recognition accuracy and risk boundary awareness, demonstrating significant advantages in overall understanding of the payment task chain, payment security boundaries, and engineering implementation complexity. This will be the key infrastructure for achieving a commercial closed loop in AI-native applications.

Low-code Access

Commercialization Barriers Are Approaching Zero

"We have componentized all the payment solutions, making them ready to use out of the box," said Zhu Lin, General Manager of Alipay's AI Payment. "In the commercialization scenarios of large models, model manufacturers can apply Alipay's capabilities as needed, allowing them to focus more on improving model performance and optimizing underlying computing power."

Through low-code access, different billing models such as subscription, pay-as-you-go, and refueling packages can be flexibly combined like building blocks. This AI payment product not only significantly reduces R&D costs for large model companies and improves payment conversion rates but also becomes a powerful driver for business growth.

New Payment Paradigm

Will Become the "Standard Foundation" for Large Model Commercialization

When the number of Agents grows exponentially and Tokens become the hard currency of the AI economy, the "payment paradigm shift" has just begun.

The three-layer architecture of Token Pay addresses the core pain points of AI commercialization. The settlement layer ensures that Agents do not "lose power" during task execution; the management layer helps large model manufacturers purchase and manage Tokens for enterprises in one go, creating a significant "efficiency lever"; the open layer lowers the access threshold to an extremely low level, allowing different billing models to be flexibly combined like building blocks, accelerating the commercialization process.

It was revealed that Alipay will further explore innovative models such as pay-per-performance and dynamic pricing, and support unified settlement across Agent frameworks. Previously, Alipay had announced a full-stack AI-native payment product system, supporting 300 million AI intelligent entities and 95% of general intelligent entity frameworks, becoming the world's first large-scale commercial AI-native payment infrastructure.