Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Hasan Toor Profile picture
@hasantoxr
Nov 15, 2023 • 13 tweets • 7 min read • Read on X
Scrolly
Internet has more than 1.6 billion websites.

But 98% don't know the best ones.

10 Secret websites that cost you nothing, but will save you hundreds of hours of your life: Image
1. Sheetplus

Sheetplus @getsheetplus will help you write Google Sheets and Excel formulas 10x faster with it's inbuild AI Tool.

Save your time and become a spreadsheet master with this amazing AI Tool.

sheetplus.ai
2. Nero AI

Use this AI enlarge, improve and clean up your photos.

This tool helps you increase image resolution without losing quality.

ai.nero.com/image-upscaler
3. Poised AI Tool

This AI tool will helps you to improve your public speaking skills.

Best tool for digital workplaces to win your next presentation, interviews and customer call.

poised.com
4. Magicstudio

Remove unwanted things in seconds with Magicstudio.

Upload, mark the bit you need removed, download.

Free to use. No signup required!

magicstudio.com/magiceraser
5. Assembly AI

Use this AI to convert audio and video files and live audio streams to text.

This AI tool helps you to multiply your existing video/ audio into text content.

Also export your transcript for subtitles and closed captions.

assemblyai.com
6. AI Image Enlarger

This AI Tool helps you enlarge, improve, and clean up your photos.

You can also enhance image resolution with this tool without reducing the quality.

imglarger.com
7. Synthesia

Use this AI to create professional videos from text.

You can create AI avatars and voices in more than 65 languages, also save you time and money.

synthesia.io
8. Quillbot

Quillbot is a paraphrasing AI website that rewrites everything and fix your grammar mistakes and write more clearly for FREE.

quillbot.com
9. Supermeme AI

Use this AI-powered meme generator to create your custom memes.

supermeme.ai
10. Texti AI

Use this AI to get instant answers to your questions while you browse the internet.

This tool helps you search smarter on Google.

texti.app
Special Bonus:

My Ultimate GPT-4 Mastery Course:

• 80+ Chapters
• 700+ New GPT-4 Prompts
• 1000+ New AI tools

Here is the free link:
hasantoxr.gumroad.com/l/gptc
Image
That's a wrap!

If you enjoyed this thread:

1. Follow me @hasantoxr for more of these
2. RT the tweet below to share this thread with your audience

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Hasan Toor

Hasan Toor Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @hasantoxr

Hasan Toor Profile picture
Apr 30
China just open-sourced a trillion-parameter model that burns fewer tokens than your favorite "efficient" US model.

Ling-2.6-1T is now public, inspectable, and benchmarkable.

The closed-model moat just got smaller.
Ant Group dropped this as a flagship, not a research toy.

1T parameters. Non-reasoning architecture. Fast-thinking by design.

It's not built to impress you with long chains of thought.

It's built to finish the task in fewer tokens than the models you're currently paying for.
The core obsession here is useful intelligence per token.

Most frontier models burn tokens narrating their thinking before they do anything.

Ling-2.6-1T skips the theater and goes straight to execution, which is the part that actually moves work forward in production.
Read 7 tweets
Hasan Toor Profile picture
Apr 28
DeepSeek V4 just went live on ZenMux with free versions at launch.

Same coding power as Claude Opus 4.7.
7x cheaper on Pro. 90x cheaper on Flash.
1M native context. MIT licensed.

Here's how to swap it into Claude Code in 3 minutes 👇
First, the numbers everyone's freaking out about.

Claude Opus 4.7 output: $25/M
DeepSeek V4-Pro output: $3.48/M
DeepSeek V4-Flash output: $0.28/M

SWE-bench Verified:
→ Opus 4.7: 80.8%
→ V4-Pro: 80.6%

Tied on coding. Tiny fraction of the bill. Image
The architecture that makes it possible.

1.6T total parameters. Only 49B activated per token.
Compute requirements: 27% of the previous generation.
KV cache: slashed to 10%.

This is efficient MoE shipped at frontier scale.
Read 14 tweets
Hasan Toor Profile picture
Apr 27
Wow... A YC-backed startup just turned game development into a single text box.

It's called CodeWisp. Type what you want and it gives you a playable game right in your browser.

No Unity. No Godot. No 5 years of tutorials. Just describe and play.

100% browser-based.
CodeWisp is a browser-based AI game builder backed by Y Combinator.

You describe the game you want in plain English.

It generates the complete code, structure, and assets automatically.

2D games. 3D games. Multiplayer browser games. All from a single prompt.
Here's how the workflow actually runs:

→ Open the browser editor (no download, no install)
→ Describe your game: mechanics, enemies, physics, levels, visuals
→ CodeWisp generates it instantly
→ Prompt edits to refine anything
→ Publish with a shareable link in one click

That's it. That's the whole process.
Read 8 tweets
Hasan Toor Profile picture
Apr 24
Ok this feels like cheating.

AntLingAGI dropped a 1T parameter model that runs like it's 7B.

No reasoning-model delay. No 40-second thinking spiral. Just instant answers at frontier scale.

Free on OpenRouter starting tonight for a full week.

Here's what I found after testing it ↓Image
First thing I noticed: token efficiency is wild.

Most 1T-class models burn through context like they're trying to lose a bet. Ling-2.6 gets to the answer without the usual 800-token preamble about what it's "about to do."

Feels built by people who actually use these models. Image
1M context window.

I threw an entire repo at it. No degradation at depth. Pulled specifics from the middle of the context without the usual "lost in the middle" collapse.

This alone makes it worth testing if you work with large codebases or long docs. Image
Read 8 tweets
Hasan Toor Profile picture
Apr 18
Google has a recording of every search you've ever made.

Every place you've ever been. Every YouTube video you've ever watched.

Go to right now.

You'll find searches from 2015. Voice recordings. GPS coordinates.

All stored. All linked to your name.

Here's how to see it and delete it:myactivity.google.com
This isn't a conspiracy theory.

A peer-reviewed study from Trinity College Dublin found that your Android phone contacts Google's servers every 4.5 minutes.

Even when you're not touching it. Even when the screen is off.

It sends your device ID, your phone number, your SIM serial number, and your location.

Even if you never signed into a Google account.

Source: Professor Douglas Leith, Trinity College Dublin, 2021.
How much data does Google have on you?

People who downloaded their full Google data reported files of 30, 60, even 150+ gigabytes.

That's the equivalent of tens of thousands of books. All about one person. You.

Go to . Download yours. See for yourself.

Warning: it can take hours to process. That should tell you something.takeout.google.com
Read 22 tweets
Hasan Toor Profile picture
Apr 17
Ok this is kind of wild.

A mystery 100B model just appeared at the top of OpenRouter out of nowhere.

No model card. No announcement. No idea which lab made it.

It's called Elephant Alpha and it's already beating half the paid models on the leaderboard. Image
The specs are what make it weirder.

→ 100B parameters
→ 256K context window
→ 32K output tokens
→ Function calling + structured output + prompt caching
→ $0 per million input tokens
→ $0 per million output tokens

You could run this against a full codebase today for free.
What actually sets it apart is the efficiency angle.

Most 100B models burn tokens like it's 2023. Elephant is tuned the other way, same answer, fewer tokens, lower latency.

Translation: cheaper agents, faster code completion, less context bloat on long docs.

This is the part nobody's screenshotting.Image
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(