🚀 Meet QMD : Quick Markdown

#markdown #python #chatgpt #llm

Hey Developers! 👋

TL;DR: Single-character syntax = significant fewer tokens = cheaper AI costs 🎯

What if I told you there's a way to make your Markdown few percentage more efficient for LLMs while keeping it just as readable for humans?

Sounds too good to be true? Well, Quick Markdown (QMD) is here to change the game! 🎮

🤔 What is Quick Markdown?

Quick Markdown (QMD) is a lightweight markup language that maintains Markdown's readability while significantly reducing storage requirements and token consumption.

Enter: QMD 🎯

🧠 Built for LLMs(in addition)
💾 Reduces token & storage usage
📏 Compact, clean, and context-aware
🔁 Still plays nicely with traditional Markdown!

It’s like Markdown, but on a diet 💪🍃

💭 Show us example, man

Yes, yes, here are examples by comparison

Traditional:    ## Heading Two        (3 tokens)
QMD:            >Heading Two          (1 token) 🎉

Traditional:    **bold text**         (4 tokens)  
QMD:            *bold text*           (2 tokens) ⚡

Traditional:    - [x] Done task       (6 tokens)
QMD:            + Done task           (2 tokens) 🍀

💵 Real impact

Comparison: Traditional vs QMD

Traditional Markdown (231 bytes):

## Shopping List

**Important** items to buy:

- Milk
- Bread
  - Whole wheat
  - Sourdough
- ~~Eggs~~ Already have

### Tasks
- [x] Buy groceries
- [ ] Call store
- [~] Check prices

### Notes

Check https://store.com for *deals*.

QMD Version (217 bytes, approximately 6.1 % reduction):

>Shopping List

*Important* items to buy:

• Milk
• Bread
  • Whole wheat
  • Sourdough
• ~Eggs~ Already have

-Tasks
+ Buy groceries
. Call store
* Check prices

-Notes

Check https://store.com for /deals/.

For larger files, the compression ratio between QMD and MD should be significantly better