| Name | qwentastic JSON |
| Version |
0.1.0
JSON |
| download |
| home_page | None |
| Summary | A simple interface for running Qwen locally |
| upload_time | 2025-02-17 00:19:17 |
| maintainer | None |
| docs_url | None |
| author | Your Name |
| requires_python | >=3.8 |
| license | MIT |
| keywords |
qwen
ai
language model
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# Qwentastic 🚀
A powerful yet simple interface for running Qwen locally. This package provides an elegant way to interact with the Qwen 1.5 14B model through just two intuitive functions.
## 🌟 Features
- **Simple One-Liner Interface**: Just two functions to remember
- `qwen_data()`: Set context and purpose
- `qwen_prompt()`: Get AI responses
- **Efficient Model Management**:
- Singleton pattern ensures model loads only once
- Automatic resource management
- State persistence between calls
- **Smart Memory Handling**:
- Optimized for both CPU and GPU environments
- Automatic device detection and optimization
- **Production Ready**:
- Thread-safe implementation
- Error handling and recovery
- Detailed logging
## 📦 Installation
```bash
pip install qwentastic
```
## 🚀 Quick Start
```python
from qwentastic import qwen_data, qwen_prompt
# Set the AI's purpose/context
qwen_data("You are a Python expert focused on writing clean, efficient code")
# Get responses
response = qwen_prompt("How do I implement a decorator in Python?")
print(response)
```
## 💻 System Requirements
- Python >= 3.8
- RAM: 16GB minimum (32GB recommended)
- Storage: 30GB free space for model files
- CUDA-capable GPU recommended (but not required)
### Hardware Recommendations
- **CPU**: Modern multi-core processor
- **GPU**: NVIDIA GPU with 12GB+ VRAM (for optimal performance)
- **RAM**: 32GB for smooth operation
- **Storage**: SSD recommended for faster model loading
## ⚡ Performance Notes
First run will:
1. Download the Qwen 1.5 14B model (~30GB)
2. Cache it locally for future use
3. Initialize the model (may take a few minutes)
Subsequent runs will be much faster as the model is cached.
## 🔧 Advanced Usage
### Custom Temperature
```python
from qwentastic import qwen_data, qwen_prompt
# Set creative context
qwen_data("You are a creative storyteller")
# Get more creative responses with higher temperature
response = qwen_prompt(
"Write a short story about a robot learning to paint",
temperature=0.8 # More creative (default is 0.7)
)
```
### Memory Management
The package automatically handles model loading and unloading. The model stays in memory until your program exits, optimizing for repeated use while being memory-efficient.
## 🤝 Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
## 📝 License
MIT License - feel free to use this in your projects!
## ⚠️ Important Notes
- First run requires internet connection for model download
- Model files are cached in the HuggingFace cache directory
- GPU acceleration requires CUDA support
- CPU inference is supported but significantly slower
## 🔍 Troubleshooting
Common issues and solutions:
1. **Out of Memory**:
- Try reducing batch size
- Close other GPU-intensive applications
- Switch to CPU if needed
2. **Slow Inference**:
- Check GPU utilization
- Ensure CUDA is properly installed
- Consider hardware upgrades for better performance
## 📚 Citation
If you use this in your research, please cite:
```bibtex
@software{qwentastic,
title = {Qwentastic: Simple Interface for Qwen 1.5},
author = {Your Name},
year = {2024},
url = {https://github.com/yourusername/qwentastic}
}
Raw data
{
"_id": null,
"home_page": null,
"name": "qwentastic",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "qwen, ai, language model",
"author": "Your Name",
"author_email": "Your Name <your.email@example.com>",
"download_url": "https://files.pythonhosted.org/packages/ab/a8/4c4cc982072420e230994c6777f68c1a6ef4361364b530d84e540af20e29/qwentastic-0.1.0.tar.gz",
"platform": null,
"description": "# Qwentastic \ud83d\ude80\n\nA powerful yet simple interface for running Qwen locally. This package provides an elegant way to interact with the Qwen 1.5 14B model through just two intuitive functions.\n\n## \ud83c\udf1f Features\n\n- **Simple One-Liner Interface**: Just two functions to remember\n - `qwen_data()`: Set context and purpose\n - `qwen_prompt()`: Get AI responses\n- **Efficient Model Management**: \n - Singleton pattern ensures model loads only once\n - Automatic resource management\n - State persistence between calls\n- **Smart Memory Handling**:\n - Optimized for both CPU and GPU environments\n - Automatic device detection and optimization\n- **Production Ready**:\n - Thread-safe implementation\n - Error handling and recovery\n - Detailed logging\n\n## \ud83d\udce6 Installation\n\n```bash\npip install qwentastic\n```\n\n## \ud83d\ude80 Quick Start\n\n```python\nfrom qwentastic import qwen_data, qwen_prompt\n\n# Set the AI's purpose/context\nqwen_data(\"You are a Python expert focused on writing clean, efficient code\")\n\n# Get responses\nresponse = qwen_prompt(\"How do I implement a decorator in Python?\")\nprint(response)\n```\n\n## \ud83d\udcbb System Requirements\n\n- Python >= 3.8\n- RAM: 16GB minimum (32GB recommended)\n- Storage: 30GB free space for model files\n- CUDA-capable GPU recommended (but not required)\n\n### Hardware Recommendations\n- **CPU**: Modern multi-core processor\n- **GPU**: NVIDIA GPU with 12GB+ VRAM (for optimal performance)\n- **RAM**: 32GB for smooth operation\n- **Storage**: SSD recommended for faster model loading\n\n## \u26a1 Performance Notes\n\nFirst run will:\n1. Download the Qwen 1.5 14B model (~30GB)\n2. Cache it locally for future use\n3. Initialize the model (may take a few minutes)\n\nSubsequent runs will be much faster as the model is cached.\n\n## \ud83d\udd27 Advanced Usage\n\n### Custom Temperature\n\n```python\nfrom qwentastic import qwen_data, qwen_prompt\n\n# Set creative context\nqwen_data(\"You are a creative storyteller\")\n\n# Get more creative responses with higher temperature\nresponse = qwen_prompt(\n \"Write a short story about a robot learning to paint\",\n temperature=0.8 # More creative (default is 0.7)\n)\n```\n\n### Memory Management\n\nThe package automatically handles model loading and unloading. The model stays in memory until your program exits, optimizing for repeated use while being memory-efficient.\n\n## \ud83e\udd1d Contributing\n\nContributions are welcome! Please feel free to submit a Pull Request.\n\n## \ud83d\udcdd License\n\nMIT License - feel free to use this in your projects!\n\n## \u26a0\ufe0f Important Notes\n\n- First run requires internet connection for model download\n- Model files are cached in the HuggingFace cache directory\n- GPU acceleration requires CUDA support\n- CPU inference is supported but significantly slower\n\n## \ud83d\udd0d Troubleshooting\n\nCommon issues and solutions:\n\n1. **Out of Memory**:\n - Try reducing batch size\n - Close other GPU-intensive applications\n - Switch to CPU if needed\n\n2. **Slow Inference**:\n - Check GPU utilization\n - Ensure CUDA is properly installed\n - Consider hardware upgrades for better performance\n\n## \ud83d\udcda Citation\n\nIf you use this in your research, please cite:\n\n```bibtex\n@software{qwentastic,\n title = {Qwentastic: Simple Interface for Qwen 1.5},\n author = {Your Name},\n year = {2024},\n url = {https://github.com/yourusername/qwentastic}\n}\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "A simple interface for running Qwen locally",
"version": "0.1.0",
"project_urls": {
"Homepage": "https://github.com/yourusername/qwentastic"
},
"split_keywords": [
"qwen",
" ai",
" language model"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "8d7a51aabe6269e6004c8d18581a65de2d5554f7f2369d05655f80c3ffbf2b3b",
"md5": "0201ca97e387a806ab9a8ccbf7e49cf4",
"sha256": "6abf16152287a939d4a1c45634deea9d6fd94c10f148d0c1ee55d9b849f85e7e"
},
"downloads": -1,
"filename": "qwentastic-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0201ca97e387a806ab9a8ccbf7e49cf4",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 5363,
"upload_time": "2025-02-17T00:19:15",
"upload_time_iso_8601": "2025-02-17T00:19:15.661900Z",
"url": "https://files.pythonhosted.org/packages/8d/7a/51aabe6269e6004c8d18581a65de2d5554f7f2369d05655f80c3ffbf2b3b/qwentastic-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "aba84c4cc982072420e230994c6777f68c1a6ef4361364b530d84e540af20e29",
"md5": "516e7238c881cd55f17a162040373aba",
"sha256": "39113f79d12fedcd4eed9522041ac8ea4ed7fe0799d70e2f049d5e0765e74543"
},
"downloads": -1,
"filename": "qwentastic-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "516e7238c881cd55f17a162040373aba",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 4628,
"upload_time": "2025-02-17T00:19:17",
"upload_time_iso_8601": "2025-02-17T00:19:17.356356Z",
"url": "https://files.pythonhosted.org/packages/ab/a8/4c4cc982072420e230994c6777f68c1a6ef4361364b530d84e540af20e29/qwentastic-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-17 00:19:17",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "yourusername",
"github_project": "qwentastic",
"github_not_found": true,
"lcname": "qwentastic"
}
