rai — your feed, without the noise. Local AI, measured.

How it works

Four gates. All on your machine.

Every tweet on your home feed gets a decision in under a second. Nothing is sent anywhere — the model runs in Ollama on localhost.

01 PREFILTER→

Obvious junk dies instantly

Engagement bait, crypto pumps, NSFW, clickbait phrasing — caught by regex before the model is even asked. A tech/business safelist protects real content from false positives.

02 LOCAL MODEL→

A 4B model scores the rest

Novelty, specificity, density, authenticity — scored by Gemma running locally (~780ms median). Concrete numbers and real experience read as signal; vague hype reads as noise.

03 THRESHOLD→

You control the strictness

Only confident noise calls get blurred. Everything is reversible — peek under any blur with one click, and tune the confidence threshold to taste.

04 CORRECTION→

Wrong calls become test cases

One tap on ✗ fixes a bad decision and records the mistake. Real corrections graduate into the eval set that gates every future version.

Measured, not vibes

Filter accuracy is a number here. Not a feeling.

Every version of the filter runs against a 109-tweet golden set before it ships — real tweets, tiered by what missing them would cost. A regression gate blocks any change that loses signal.

100%

Critical-signal recall

Job leads, paid pilots, key releases — zero lost.

96.2%

Signal recall

The metric you could never feel by scrolling.

89.3%

Noise caught

Including tech-flavored engagement bait.

Why measure at all? A feed filter has two failure modes with very different visibility. Noise that leaks through is annoying — but you see it, so you fix it. Signal that gets hidden is invisible — you never know what you lost, so every hand-tuned "improvement" quietly drifts toward over-hiding. The eval suite guards the side no human can feel, and the ✗ button turns your real corrections into test cases.

Also: YouTube

Open YouTube for music. Leave with music.

The same extension runs an inverted filter on YouTube: everything is blurred by default, and only music (and optionally motivational videos) is revealed. For when you came to press play, not to lose an evening.

Blur-by-default

The homepage can't grab you

Every thumbnail is blurred until the local model confirms it's music. The trending video you didn't come for stays a gray rectangle.

Title + channel only

Fast enough to be invisible

Classification uses just the title and channel with a 2B model, so the grid resolves in a blink. Search and channel pages are never touched.

Shorts nudge

Doom-scroll circuit breaker

A counter tracks Shorts binges. Past your limit — 10 Shorts or 5 minutes — a gentle full-screen nudge asks if this is really where the evening goes.

++++

No Ollama? No problem.

Cloud mode. Nothing to install.

Local mode is free forever and never leaves your machine. Cloud mode is for everyone who doesn't want to run a local AI model: rai runs the classification for you. While rai is in beta, it's free — no card, no subscription, no account, no email.

Install the extension

From the Chrome Web Store or GitHub — see below.

Switch Mode to Cloud

⚙ settings → Mode → Cloud. No setup, no forms.

Get a free key

One click. The key is your account — save it and scroll.

rai cloudFree beta

Same filter. Same rules. Zero setup.

~50,000 classifications preloadedmonths of normal scrolling on both platforms
One click, one keyno card, no account, no email
X and YouTube, both coveredtext and titles only — never images, never your identity

Get the extension → Privacy policy

Cloud mode sends tweet text and video titles to rai's server for classification. The privacy policy lists exactly what's sent and to whom.

Install

Five minutes, one script.

The setup script installs the local AI runtime, downloads the right model for your machine, fixes permissions, and proves the whole pipeline with a live classification before you ever open Chrome.

Get the code and run setup

git clone https://github.com/phuaky/xrai.git && cd xrai
bash scripts/setup.sh
# ...
✓ Model classified a test tweet: {"prediction":"noise","confidence":0.78}

Load the extension

Open chrome://extensions → enable Developer mode → Load unpacked → select the extension/ folder.

Open x.com

The rai pill appears bottom-right with live shown/hidden counts. 👁 peeks under any blur, ✗ corrects mistakes, ⚙ tunes everything.

Privacy & rules

Fast does not mean sneaky.

Four plain answers before you install. rai behaves like an ad blocker, not a bot.

01Does anything leave my machine?

In local mode, no. The model runs in Ollama on localhost, and there's no account, no telemetry, and no analytics — on this site either. Cloud mode sends tweet text and video titles (never images, never your identity) to rai's server; the privacy policy lists every byte.

02Is this allowed by X and YouTube?

rai reads what's already on your screen and hides some of it with CSS — the same mechanics as an ad blocker.

Never calls X or YouTube APIs — reads already-rendered DOM only
Never posts, likes, follows, or clicks anything for you
Hiding is CSS-only — blur and display, fully reversible
Reply suggestions are copy-paste only, never auto-sent

03What do I need to run it?

Local mode: a Mac with Chrome and ~2 GB of disk for the model — the setup script handles the rest. Cloud mode: just Chrome. Everything is open source on GitHub.

04What if it hides something important?

Every hidden post is one click from visible — blur, not delete. And the failure you can't see is exactly what the eval suite guards: 100% recall on critical signal, gated on every release, with your ✗ corrections feeding the test set.