Web search built for AI scale

Web infrastructure for AI to search, extract, monitor, and reason over the world's information.

Welcome to the official Parallel organization on Hugging Face. We build web APIs for AI agents. Give your models access to current information: search the web, extract useful content, research complex questions, monitor changes, and make decisions grounded in web evidence.

Website · Get started · Documentation · GitHub · Blog

Get started with your agent

Copy this prompt into Claude Code, Codex, Cursor, or your preferred coding agent:

Use curl to read https://parallel.ai/agents.md and perform the setup to install Parallel.

View the setup instructions · Start building for free

Build with Parallel

• Search API — Find relevant web sources and retrieve excerpts designed for language models.
• Extract API — Turn web pages into clean content for your agent's context.
• Decisions API — Answer closed-ended questions using evidence from the live web.
• Task API — Run deep research and return structured outputs with citations.
• FindAll API — Discover companies, people, and other entities that match your criteria.
• Monitor API — Track topics on the web and receive updates when new information appears.

Research and evaluation

We're interested in how well AI systems find evidence and use it: web search, retrieval, deep research, and grounded decision-making. This organization is a home for Parallel's evaluation datasets and research resources.

For your own evaluations, start with our Search evaluation guide and API documentation.

Resources for developers and agents

• API playground — Try Parallel and get an API key.
• MCP integration — Connect Parallel to agents through the Model Context Protocol.
• Documentation index for agents — Find API guides and integration references.
• GitHub — SDKs, examples, and developer tools.