Parallel Web Systems, Valued at $740 Million, Unveils 'Extract' API to Optimize Web Data for AI Agents

Palo Alto, California – Parallel Web Systems, the AI startup founded by former Twitter CEO Parag Agrawal, has announced the launch of its new "Parallel Extract" API, designed to efficiently retrieve and format web content for artificial intelligence agents. This new offering, part of the company's Agent Tools bundle, aims to enhance the accuracy and token efficiency of AI systems interacting with the vast expanse of the internet. The announcement follows a significant Series A funding round that valued the company at $740 million.

The Parallel Extract API functions by fetching all content from a given URL and returning it in markdown format. Users can choose between full detail or a compressed version, optimizing for better token efficiency in AI applications, as stated by Parallel Web Systems. > "Today, we're launching Parallel Extract, a new API in our Agent Tools bundle. When given a URL, Extract fetches all content from that page and returns it in markdown, either in full detail or in a compressed form for better token efficiency," the company announced on social media.

This tool is poised to benefit a range of AI-driven tasks, including the extraction of coding documentation, processing of PDF research papers, summarization of news articles without extraneous elements, and analysis of financial filings. By providing clean, structured data, Parallel Extract aims to reduce "hallucinations" and improve the reliability of AI models. The API leverages the same proprietary index and retrieval infrastructure that powers Parallel's existing Task and Search APIs, creating a synergistic suite for AI agents.

Founded in 2023 by Parag Agrawal, Parallel Web Systems operates with a core mission to adapt the World Wide Web for artificial intelligences. Agrawal believes that AI agents will soon become the primary users of the internet, necessitating new infrastructure for web data access and computation. The company's offerings are built to enable AIs to perform high-value tasks using web data more effectively than traditional search methods.

The company recently secured $100 million in a Series A funding round, co-led by Kleiner Perkins and Index Ventures, with participation from Khosla Ventures and other existing investors. This investment brings Parallel Web Systems' valuation to $740 million post-money, underscoring investor confidence in its vision. Parallel claims its technology surpasses leading AI models like GPT-5 in deep web research benchmarks, offering superior accuracy and cost-effectiveness for AI-native web search.

Parallel Web Systems plans to use the new capital to accelerate product development and customer acquisition, while also addressing challenges like paywalls and login barriers that restrict AI access to web content. Agrawal has indicated plans to develop an "open market mechanism" to incentivize publishers to keep their content accessible to AI systems, aiming to foster a more programmatic web for AI.