For Data Buyers

Portex supports two primary ways for purchasing datasets with USDC: through our user interface and via our Data Agent. Each approach offers unique benefits, catering to different buyer preferences and technical requirements.

Purchasing Via the User Interface

Inspired by the success of platforms like the Hugging Face data portal and the Kaggle Data Explorer, our user interface allows buyers to browse and purchase datasets with ease. Users can explore datasets under various categories, viewing summary statistics such as dataset schemas, sample distributions, and size. Once they’ve identified a dataset of interest, buyers can finalize the purchase onchain using USDC, after which a secure download link is provided for immediate retrieval. The interface is designed to simplify discovery and encourage experimentation, making data more accessible to both novices and expert analysts.

Purchasing Via the Data Agent

For more automated or programmatic interactions, Portex offers a Data Agent — an agentic API that enables buyers (including human operators and autonomous agents) to access the marketplace. Rather than clicking through a graphical interface, buyers can submit queries in natural language or concise instructions specifying the desired data. The Data Agent then parses the request, provides pricing information, and coordinates the USDC-based purchase transaction onchain.

Depending on the volume of data purchased, the agent tailors the delivery method. Smaller datasets may be transmitted directly via an API response, while larger datasets are supplied through a secure download link, available immediately upon confirmation of payment. This flexible delivery mechanism empowers AI systems, automated workflows, and other high-velocity buyers with near real-time access to the data they need to operate effectively.