# Revenue model ## **Transaction Fees** GPGPU is a decentralized GPU cloud service that intermediates GPU resource transactions between providers and users. GPU resources are used for various purposes, such as AI service development, data processing, deep learning, and machine learning. Through these transactions, GPGPU earns transaction fees. A certain percentage fee is charged for each transaction, with payment options including credit cards and cryptocurrencies. The transaction fee is charged at the completion of the transaction, with 1.5% deducted from both the provider and the user. ## **Period Contract Margin** Users can enter into period contracts with GPU providers for a stable environment. During the contract period, GPU providers are obligated to provide consistent GPU resources, allowing users to establish an environment akin to owning their own GPUs. The cost of the period contract is based on the hourly rental cost of the GPU devices, with potential discounts negotiated between the provider and the user depending on the duration and scale of usage. GPGPU also charges transaction fees for period contracts, applying lower rates for larger transactions. ## **Rental Fees** GPGPU can enter into rental agreements directly with GPU suppliers or purchase GPU devices to form a GPU pool. This allows GPGPU to generate rental income from its own GPU pool rather than merely brokering transactions. If rental agreements are made with GPU suppliers, GPGPU earns a margin on the difference between the rental cost paid to the supplier and the usage fee paid by the user. Additionally, direct operation or period contracts enable GPGPU to provide enterprise-grade services, leading to higher rental income margins. ## **Subscription Fees** Users can choose a subscription model that offers the advantages of period contracts with more flexible pricing. While similar to period contracts, subscriptions have key differences: users can start and stop subscriptions as needed and freely change their subscription plans. Monthly or annual subscriptions are available, and users can upgrade their plans mid-term for extended usability. This model is beneficial for medium-term stable service usage with flexible scalability. ## **Discounts** Discount rates are applied based on the scale and duration of GPU rentals. These discounts provide users with various options, incentivizing active and long-term usage. A maximum discount of 20% is available, divided equally between rental scale and duration. The rental duration offers a maximum discount up to 12 months, while the rental scale applies a discount up to a maximum cost of $10,000. If the projected cost of GPU resources exceeds $10,000 over 12 months, the maximum discount of 20% is applied. Additionally, users can negotiate further discounts with GPGPU for large-scale usage. By offering competitive transaction fees, period contract margins, rental fees, subscription fees, and attractive discounts, GPGPU provides a cost-effective and flexible solution for both GPU providers and users, fostering a sustainable and efficient AI and computing ecosystem. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://gpgpu.gitbook.io/gpgpu/overview/revenue-model.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.