---
title: Huihui Qwen3.8 27B Abliterated GGUF
emoji: ⚡
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 6.25.0
python_version: '3.12'
app_file: app.py
pinned: false
short_description: Chat demo for Huihui Qwen3.8 27B Abliterated GGUF
---

# Huihui Qwen3.8 27B Abliterated (GGUF) Demo

Conversational chat demo running [**huihui-ai/Huihui-Qwen3.8-27B-abliterated-GGUF**](https://huggingface.co/huihui-ai/Huihui-Qwen3.8-27B-abliterated-GGUF) using `llama.cpp` on Hugging Face ZeroGPU.

## Features

- **Fast Inference**: Uses GGUF quantization with GPU offloading via `llama-cpp-python`.
- **ZeroGPU Acceleration**: Dynamic GPU allocation on NVIDIA GPUs.
- **Streaming Responses**: Real-time response streaming into a modern bubble chatbot UI.
- **Configurable Generation**: Customizable system prompt, temperature, top-p, top-k, repetition penalty, and max tokens.
- **MCP Server Ready**: Built-in Model Context Protocol server support (`mcp_server=True`).

## Notice

This model has significantly reduced safety refusal filtering. It may generate sensitive or uncensored content. You are responsible for adhering to applicable laws and policies.