Futures
Access hundreds of perpetual contracts
TradFi
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Pre-IPOs
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
AI iteration is moving too fast: just learned Headless Chrome, and now the new Rust star Obscura has fully surpassed it
A few days ago, I was still feeling: Headless Chrome is slow to start, high memory usage, easy to be detected by anti-crawling measures, and when AI Agents run at high concurrency, server resources become severely strained.
Then suddenly on GitHub, a lightweight headless browser engine written in Rust appeared—Obscura.
It is specifically designed for AI Agents and modern web crawlers, running the true V8 JavaScript engine, supporting Chrome DevTools Protocol (CDP), and can directly serve as a drop-in replacement for Puppeteer and Playwright.
The core motivation of the author is simple and direct: because Chrome is too heavy and too slow.
How does Obscura solve Chrome’s longstanding issues?
For those of us involved in automation, writing Agents, and large-scale scraping, the pain points of Headless Chrome are well known:
🔸High memory consumption: a single instance often over 200MB, putting huge pressure on servers at high concurrency.
🔸Slow startup and loading: about 2 seconds to start, dynamic pages often take 500-800ms to load.
🔸Large size: requires a full Chrome + Node.js environment, making deployment inconvenient.
🔸Weak anti-detection capabilities: fingerprinting is easy to expose, requiring additional plugins to maintain.
Obscura provides highly competitive answers to these problems (data sourced from the project’s GitHub README and early testing):
🔸Memory usage: about 30MB (compared to Chrome’s 200MB+)
🔸Binary size: about 70MB for a single file, zero additional dependencies
🔸Page load time: static pages around 51ms, dynamic JS/XHR/React sites around 80-85ms
🔸Startup speed: nearly instant
🔸Compatibility: fully supports CDP, can seamlessly switch between Puppeteer/Playwright, and existing scripts require little to no modification
It’s not a simplified “HTML-only rendering” tool, but a true embedding of the V8 engine, capable of perfectly handling modern SPA sites, while significantly cutting down the bulky parts of Chromium.
Why is the Rust + V8 combination so efficient?
Rust’s memory safety, no-GC pauses, and high concurrency capabilities make Obscura perform excellently in resource control. It embeds V8 directly into the binary, avoiding the overhead of traditional browser process communication, and the entire engine is designed to be minimal and focused.
Another practical highlight is the built-in Stealth mode:
🔸Randomize browser fingerprint
🔸Proactively intercept trackers
🔸Significantly reduce the risk of detection by anti-crawling systems
This is especially friendly for AI Agents—when the Agent needs real-time browsing and decision-making, there’s no need to stack a bunch of anti-detection plugins, improving stealth and stability.
Getting started: surprisingly simple
Obscura has a very low barrier to entry, supporting cross-platform (Linux, macOS, Windows):
🔸CLI single-page scraping: obscura scrape
🔸Service mode: obscura serve --port 9222, then point existing Puppeteer/Playwright scripts to this port
🔸Batch processing: supports multiple URLs concurrently, returning structured JSON
Installation is also very easy—just download the corresponding binary for your platform from GitHub Releases, no need to install Chrome or Node.js.
GitHub address: