Parallel Agents allows users to run multiple /agent queries at the same time, streamlining data extraction from various sources. The service is designed to handle a range of data needs, from extracting a single data point to compiling extensive datasets. Users can specify what data they want, and the system manages the extraction process. Firecrawl’s Agent can pull information from websites like Y Combinator, Crunchbase, and GitHub, delivering structured results in JSON format. Use cases for the Agent are diverse, catering to different sectors such as lead generation, e-commerce, market research, and real estate. For example, users can request lists of companies or product listings effortlessly. The integration is straightforward with support for APIs, SDKs, and command-line tools, making it accessible for developers. Firecrawl provides a sample code for various programming languages, illustrating how to configure and initiate data extraction. Pricing is based on the complexity of queries, with five free runs available daily during the research preview. Simple queries consume fewer credits, while more complex ones use more. Firecrawl offers different model options for queries, including Spark 1 Fast for quick lookups and Spark 1 Pro for more intricate tasks. The system's intelligent waterfall process optimizes performance by choosing the best model for the job. FAQs address common inquiries about usage and pricing, ensuring potential users have the information they need. Overall, Firecrawl's Agent is positioned as a powerful tool for effective data extraction across multiple industries.