Why Chrome Eats Your RAM & How We Built a Manifest V3 Extension to Slash Memory Usage by 95%

Sudeep S Nair — Sat, 06 Jun 2026 05:14:55 +0000

If you are reading this, your browser is probably consuming several gigabytes of RAM right now.

We’ve all been there: you’re in the middle of a coding session, Notion is open, Slack is running, you have 15 documentation tabs, 4 StackOverflow threads, and 3 Figma files. Suddenly, your IDE starts lagging, your MacBook's fans kick in, and your battery percent drops like a stone.

You open Chrome’s Task Manager, and there it is: Google Chrome is eating 12GB of RAM.

In this post, we’re going to dive into the technical reasons why Chrome is such a memory hog, compare legacy Manifest V2 extensions with the modern Manifest V3 architecture, and look at how we built a lightweight tab suspension extension to cut Chrome's memory footprint by up to 95%.

The Root Cause: Process Isolation & Sandboxing

To understand why Chrome uses so much memory, we have to look at its underlying architecture. Chrome operates on a multi-process architecture.

Instead of running the entire browser in a single process, Chrome spawns separate processes for:

The browser shell (UI, navigation, bookmarks)
The GPU process (rendering graphics)
Every single tab you open
Every extension you run

Here is a visual represention of how Chrome maps memory:

graph TD
Chrome[Chrome Browser Process] --> GPU[GPU Process]
Chrome --> Tab1[Tab 1: Slack - 800MB]
Chrome --> Tab2[Tab 2: Figma - 1.2GB]
Chrome --> Tab3[Tab 3: GitHub - 150MB]
Chrome --> Ext1[Extension: Legacy Session Saver - 250MB]
Chrome --> Ext2[Extension: Ad Blocker - 180MB]

This design is brilliant for security and stability. If a memory leak or script error crashes a website in Tab 1, it won't crash Tab 2 or the browser shell.

However, process isolation comes with a massive overhead. Each process requires its own memory allocation, its own engine instance (V8), and its own utility scripts. If you have 40 tabs open, you are running 40 isolated mini-browsers in the background.

Legacy Manifest V2 vs. Manifest V3 Resource Management

For years, developers relied on tab managers and ad blockers to optimize their browsers. However, many of these legacy extensions are built on Manifest V2 (MV2).

MV2 extensions run persistent background pages. These are invisible HTML documents running scripts in the background 100% of the time, even if you haven't clicked the extension icon all day. This causes chronic background memory drain.

Enter Manifest V3 (MV3)

Google’s Manifest V3 replaces persistent background pages with event-driven Service Workers.

Unlike MV2, an MV3 Service Worker only loads when it needs to respond to a specific browser event (like opening a new tab or saving a link). Once the script finishes executing, Chrome completely terminates the service worker process, freeing its memory.

Here is a simplified comparison of background resource usage:

javascript

// LEGACY (Manifest V2) background.js - Persists in memory forever
let savedTabs = [];
chrome.tabs.onCreated.addListener((tab) => {
savedTabs.push(tab);
// This process never unloads. If a memory leak occurs here, it remains active.
});
// MODERN (Manifest V3) service-worker.js - Event-driven, sleeps automatically
chrome.tabs.onCreated.addListener(async (tab) => {
// Service worker wakes up, writes tab metadata to local storage
const data = await chrome.storage.local.get("tabs") || { list: [] };
data.list.push(tab);
await chrome.storage.local.set(data);
// Service worker unloads from RAM within seconds of completion
});

Transitioning your tools to MV3 is one of the most effective ways to find a permanent chrome extension memory leak fix.

The Solution: Active Tab Suspension

Since we can't change Chrome's process isolation architecture, the only way to reclaim memory from background tabs is to terminate their renderer processes. This is called tab suspension.

Tab suspension works by:

Monitoring which tabs are inactive.
Capturing the state of the tab (URL, title, scroll position, and history).
Replacing the tab's active process with a lightweight placeholder page.
When you click back onto the tab, restoring it exactly where you left it.
By suspending an inactive tab running a heavy application like Figma, you can reduce its memory footprint from 1GB+ down to less than 15MB.

How We Built Tabnxt to Solve Browser Bloat

We got tired of sluggish systems, so we built Tabnxt — an AI-powered, Manifest V3 tab manager designed specifically to reduce Chrome RAM usage proactively.

Here is how we optimized it for maximum performance:

1. Real-Time Memory Dashboard
Tabnxt queries Chrome's internal processes to show you exactly how many megabytes of RAM each tab is chewing up. No more opening Chrome's Task Manager; you can see the resource hogs instantly in your sidebar.

2. Event-Driven Auto-Suspension
Using Manifest V3 event listeners, Tabnxt automatically puts background tabs to sleep after a customizable period of idle time. It uses zero persistent memory when you aren't active.

3. Smart Workspace Management
Instead of keeping 50 tabs open "just in case," Tabnxt lets you save browser sessions and group them into custom workspaces. This lets you close the tabs entirely, releasing 100% of their RAM, while keeping your session safe.

The Results: Real RAM Optimization
When we benchmarked Tabnxt against legacy tools in a 40-tab workflow, the difference was staggering:

Without Tabnxt: 7.4 GB RAM consumed
With Tabnxt (Auto-Suspend enabled): 650 MB RAM consumed
If you're dealing with browser overload, check out our guide on how to handle too many tabs open in Chrome, or see how Tabnxt stacks up against legacy tools in our Tabnxt vs Session Buddy comparison.

We are currently completing the private beta of Tabnxt (launching in the Chrome store next week). If you want to stop browser memory bloat and save your laptop's battery, you can join our free pre-launch waitlist here.

Have questions about Chrome's memory management or building Manifest V3 extensions? Let's discuss in the comments below!

DEV Community: Sudeep S Nair

Why Chrome Eats Your RAM & How We Built a Manifest V3 Extension to Slash Memory Usage by 95%