README.md

---
title: Qwen API
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: docker
app_file: app.py
pinned: false
license: apache-2.0
tags:
  - qwen
  - uncensored
  - llama-cpp
  - gguf
suggested_hardware: a10g-small
---

# Qwen3.5-9B Uncensored API Interface

API interface for [HauhauCS/Qwen3.5-9B-Uncensored-HauhauCS-Aggressive](https://huggingface.co/HauhauCS/Qwen3.5-9B-Uncensored-HauhauCS-Aggressive).

## Features

- 9B parameters with 262K context window
- Fully uncensored (0/465 refusals)
- Multimodal capable (text, image, video)
- Supports 201 languages
- Q4_K_M quantization via llama.cpp

## API Usage

### Python

```python
from gradio_client import Client

client = Client("Ngixdev/qwen-api")

result = client.predict(
    prompt="Your question here",
    system_prompt="You are a helpful assistant",
    temperature=0.7,
    top_p=0.8,
    max_tokens=1024,
    api_name="/api_generate"
)
print(result)
```

### cURL

```bash
curl -X POST https://ngixdev-qwen-api.hf.space/api/api_generate \
    -H "Content-Type: application/json" \
    -d '{
        "data": [
            "Your question here",
            "You are a helpful assistant",
            0.7,
            0.8,
            1024
        ]
    }'
```

## Parameters

| Parameter | Type | Default | Description |
|-----------|------|---------|-------------|
| prompt | string | required | User prompt/question |
| system_prompt | string | "" | System instruction |
| temperature | float | 0.7 | Sampling temperature (0.0-2.0) |
| top_p | float | 0.8 | Nucleus sampling (0.0-1.0) |
| max_tokens | int | 1024 | Maximum tokens to generate |