20q / README.md
david-ar's picture
Upload folder using huggingface_hub
b584b09 verified

A newer version of the Gradio SDK is available: 6.12.0

Upgrade
metadata
title: TwentyQ
emoji: 🧠
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 6.9.0
app_file: app.py
pinned: false
license: mit
models:
  - david-ar/20q
short_description: The world's smallest chat model  187K params, 214KB

TwentyQ — The World's Smallest Chat Model

A 2-bit quantized neural network that plays Twenty Questions. Think of any object and the model will try to guess it by asking up to 20 yes-or-no questions.

Parameters 187,200
Precision 2-bit (Q2_0)
Model size 187 KB weights + 27 KB vocab
Architecture Single-layer associative network
Training data ~75M human conversations
Context window 20 turns
Output classes 1,200

Usage

Loaded with the standard transformers API:

from transformers import pipeline

pipe = pipeline("text-generation", model="david-ar/20q", trust_remote_code=True)

Valid Responses

  • First question: Animal, Vegetable, Mineral, Other
  • Regular questions: Yes, No, Probably, Doubtful, Maybe, Unknown
  • Guesses: Yes, No, Close