README.md

metadata

title: SuperGemma4-26B Uncensored
emoji: 🚀
colorFrom: purple
colorTo: pink
sdk: gradio
sdk_version: 6.16.0
app_file: app.py
pinned: false
license: apache-2.0
python_version: 3.11

SuperGemma4-26B Uncensored (GGUF - CPU)

Powerful 26B parameter uncensored AI model running on CPU using GGUF format.

Features

• 💬 Interactive chat without restrictions
• 💻 Code generation (Python, JavaScript, Java, C++, etc.)
• 📝 Uncensored text generation
• 🔥 26B parameters with Q4_K_M quantization
• ⚡ Runs on CPU - No GPU required!

Model

• Base: Jiunsong/supergemma4-26b-uncensored-gguf-v2
• Parameters: 26 billion
• Quantization: Q4_K_M (4-bit)
• Type: Uncensored
• Format: GGUF (llama.cpp compatible)

Hardware

This Space runs on CPU - no GPU required! Uses llama.cpp for efficient CPU inference.

Performance

• Speed: ~2-5 tokens/second on CPU
• Memory: ~16GB RAM
• Context: 4K tokens

Usage

Web Interface

Use the tabs above to chat, generate code, or create text.

API

from gradio_client import Client

client = Client("sagarleet/test")
result = client.predict("Write a Python function", 500, 0.7, 0.9, 40, api_name="/predict")
print(result)

Disclaimer

This is an uncensored AI model. Use responsibly and ethically.