3. View the mindmap

Copy the code generated by Gemini.

Open markmap.js.org/repl and paste it in the left-hand column.

You can interact with your mindmap and download it in SVG or HTML format.

2. Enter the prompt

Copy/paste the prompt from my own conversation:
g.co/gemini/share...

Gemini will generate code that will allow you to view your mindmap.

1. Upload your book

Open Gemini and upload your book in PDF format.

You can also use Google AI Studio by selecting the 1.5 Pro model.

Gemini can turn any book into an interactive mindmap

You can do it for books of hundreds of pages thanks to the massive context window and with a single prompt.

Perfect for those who memorize better visually (demo and prompt below):

Reasoning models are becoming the norm

So a friendly reminder that OpenAI has released a guide to writing prompts for them.

Few important points and example:

Google has just released a very good reasoning model... and it's free 🔥

Already the best model on Chatbot Arena.

Select "Gemini 2.0 Flash Thinking" in AI Studio and you can start using it right away.

Based on Flash so VERY fast.

Reasoning models are becoming increasingly popular

But I personally don't see the point of paying $200/mo for o1 Pro.

No doubt some people will have a use for it but for the majority Qwen QwQ is a better choice:

Meta has just released Llama 3.3 70B which is more powerful than GPT-4o and 25x cheaper.

Yes. 70B and better than GPT-4o.

This model is also as powerful as the 405B version of Llama 3.1.

Open source is really winning at every level.

Closed AI models are no longer dominant.

You can use free and open source alternatives for all of them:

Text generation
GPT-4o → Nvidia 3.1 Nemotron

Coding
Sonnet 3.5 → Qwen 2.5 Coder

Reasoning
o1-preview → QwQ

Multimodal tasks
Gemini 1.5 Pro → Pixtral

Access them 🧵

I see a lot of people overspending to build full-stack apps without coding

Cursor → Replit AI
v0/Bolt → Replit Agent
Vercel → Replit Deployment

$60/mo → $15/mo

All-in-one but 4x cheaper.

Replit has just received a major update

You can clone any interface with a screenshot and a prompt.

Screenshot → Paste into Replit Agent → “Clone this app” → Done!

All that's left is to add your features, a backend and more (without coding). More below.

I built this complete app without writing a line of code. Only English.

- Replit Agent + AI + deployment
- Backend with Neon in Replit
- Log in with Google OAuth
- Subscription with Stripe
- Domain with Cloudflare

Again, not a single line written by myself.

You can build whatever you want.