{"id":920,"date":"2025-10-25T06:20:00","date_gmt":"2025-10-25T06:20:00","guid":{"rendered":"https:\/\/yipinqian.com\/?page_id=920"},"modified":"2026-07-02T06:21:59","modified_gmt":"2026-07-02T06:21:59","slug":"fatty","status":"publish","type":"page","link":"https:\/\/yipinqian.com\/ms\/fatty\/","title":{"rendered":"Fatty Singapore AI Model – A Micro Local SG Artificial intelligence –"},"content":{"rendered":"\n

Fatty Singapore AI Base Model – Beta Trial Testing Version 1.6<\/strong> – A Micro Local SG Artificial intelligence<\/h2>\n\n\n\n

 <\/p>\n\n\n\n

Fatty Singapore AI Base Model ( Artificial intelligence<\/strong> ) – Beta Trial Testing Version 1.6 – Supervised Machine Learning Model<\/strong> – Users may use the model for inference only. No fine-tuning, distillation, or derivative works allowed without written permission.<\/h3>\n\n\n\n


A 42M Parameter AI Singapore Soul in the Making Local Clean From Scratch – Build by Singaporean<\/h3>\n\n\n\n

The Fatty AI Method:<\/strong> Pure Supervised Learning
Fatty AI is built from the ground up using Supervised Machine Learning. This is the core of how it thinks, but it’s a very different process from what you might have heard from large AI companies.<\/p>\n\n\n\n

Fatty will only expand tools after the base is stable with more learning.<\/p>\n\n\n\n

What is Supervised Learning?<\/strong>
In simple terms, Supervised Learning is like teaching a child with flashcards.<\/p>\n\n\n\n

You show the child a flashcard with a Question (the input).<\/p>\n\n\n\n

On the back is the correct Answer (the desired output).<\/p>\n\n\n\n

By studying thousands of these flashcards, the child learns the underlying patterns and eventually can answer new, similar questions on their own.<\/p>\n\n\n\n

This is exactly how Fatty AI was trained + UI testing and check model response<\/p>\n\n\n\n

\n
\n
\n
\"Fatty<\/figure>\n\n\n\n
\"Fatty<\/figure>\n\n\n\n
\"Fatty<\/figure>\n<\/figure>\n<\/div>\n<\/div>\n\n\n\n

How Fatty AI Was Supervised?<\/strong><\/h4>\n\n\n\n

I acted as the teacher. For over 600,000 examples, approximates around 786 million characters in this version which is still underfed. I provided the model with a perfect pair:-<\/p>\n\n\n\n

This Model is heavily trained on routes and address and underfed on most daily use conversations on this Trial Testing Version. Estimated growth with minimum 5 million training examples just for local in 2026.<\/p>\n\n\n\n

It does help tourist to navigate for around 19 to 20 Hotels at Orchard Road, Tanglin Rd, Scotts Rd to Malls right now. Examples: How to go Tudor Court from Hyatt Hotel? How to go Yotel from Plaza Singapura? How to go Paragon? etc.<\/strong><\/p>\n\n\n\n

A precise instruction or question ( e.g., How do I walk from Concorde Hotel to Camden Medical? )<\/p>\n\n\n\n

The exact, correct answer I wanted it to learn (e.g., Starting point: Concorde Hotel\u2026 [ – detailed route – ]\u2026)<\/p>\n\n\n\n

The model’s only job was to analyze these millions of Question and Answer, pairs and discover the deep patterns that connect them. Through this intense, focused training over hundreds of epochs, it didn’t just memorize, it learned to understand the structure of navigation questions, the properties of locations, and how to generate a correct response.<\/p>\n\n\n\n

The Key Difference: Deep Education vs Surface Level Manners
This is where Fatty AI differs completely from many models.<\/strong><\/p>\n\n\n\n

\n