Documentation

Parallel Research

AI-powered research agent for deep web intelligence, lead generation, and automated outreach. Written in Rust for maximum performance and reliability.

Quick Start

Get up and running in three steps.

1

Install

Install via the curl script or build from source with Cargo.

bash
curl -sSf https://parallel.research/install.sh | sh
2

Configure

Add your API key and choose a search backend.

~/.parallel-research/config.toml
[llm]
provider = "deepseek"
api_key  = "sk-your-key"
model    = "deepseek-v4-flash"

[search]
backend  = "hybrid"
3

Run

Launch a research session with a natural language query.

bash
parallel-research run "Find B2B SaaS startups in Berlin with 10-50 employees"

Key Sections

Explore the documentation to get the most out of Parallel Research.

⚡

Quickstart

Install, configure, and run your first research session in minutes.

→
⚙️

Configuration

Full reference for config.toml: LLM, search, agents, notifications, and more.

→
🔧

Tools

49 built-in tools: web search, OSINT, browser automation, Git, file ops.

→
📡

API Reference

HTTP API, SSE event stream, and CLI commands for programmatic access.

→

What is Parallel Research?

Parallel Research is an autonomous AI agent that conducts deep web research, extracts contacts, generates leads, and automates outreach — all from a single command-line interface.

Built in Rust for speed and reliability, it orchestrates multiple sub-agents that work in parallel, using 49 tools including web search, browser automation, OSINT pipelines, and file operations.

🦀 Rust Core49 ToolsMulti-AgentOSINT PipelineTUI InterfaceHTTP APIMCP Support7 Search Backends

Running Modes

CLI

parallel-research run "your query"

Single-shot research from the terminal.

TUI

parallel-research tui

Interactive terminal UI with real-time progress.

HTTP Server

parallel-research serve

REST API with SSE streaming for integrations.