#1212th of 16 in Data and scraping for AI
moli
Headless browser in Rust for AI agents, with CDP and WebDriver support
- Stars
- 13.8k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Rust
Missed the latest GitHub check; these facts may be older.
Overview
Moli is a standalone browser kernel written in Rust (V8, Stylo, Taffy, Parley), not a Chromium wrapper. The CLI fetches pages as HTML, Markdown, JSON or a semantic tree, and `moli serve` exposes CDP, WebDriver Classic and WebDriver BiDi on one endpoint. Layout and painting run only when `--layout` is set and an operation needs them.
Who it is for: Engineers building crawlers, browser agents or RL environments
Strengths
- Single binary serves CDP, WebDriver Classic and BiDi; no ChromeDriver or Chrome install needed
- Reported median RSS of 73 MiB on a 192-URL crawl, versus 773 MiB for Chrome Headless
- CLI dumps Markdown, semantic tree text, PNG screenshots and PDF directly
- Playwright connects over CDP; Linux, macOS and Windows supported
Weaknesses
- Lexbench pass rate is 81.9% against Chrome's 99.8%, so some sites will fail
- No GUI, GPU compositor, or high-fidelity Canvas, WebGL or media playback
- Screenshot and print modes are partial; real layout needs the `--layout` flag
- Benchmarks are published by the project's own authors
What it needs
- no GPU
- Needs libcurl
- port 9222
Also in Data and scraping for AI
See all 16| Rank | Project | Score |
|---|---|---|
| 1 | mineruDocument parser that converts PDFs, images and Office files to Markdown and JSON | 80 out of 100 |
| 2 | Crawl4AIPython web crawler that converts pages to LLM-ready Markdown and JSON | 78 out of 100 |
| 3 | doclingPython library and CLI that converts PDFs, Office files and more into structured text | 77 out of 100 |
| #4 | markitdownPython tool that converts PDF, Office, HTML and audio files to Markdown for LLMs | 71 out of 100 |
| #5 | FirecrawlWeb scraping and crawling API that returns LLM-ready markdown | 67 out of 100 |