{"id":29873141,"url":"https://github.com/0x0806/unleashedllm","last_synced_at":"2026-04-19T17:02:08.070Z","repository":{"id":305884600,"uuid":"1023552073","full_name":"0x0806/UnleashedLLM","owner":"0x0806","description":"A sophisticated command-line interface tool for downloading, managing, and interacting with powerful AI models completely offline.","archived":false,"fork":false,"pushed_at":"2025-07-22T13:31:14.000Z","size":41,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-07-22T14:39:51.615Z","etag":null,"topics":["ai","aimodels","download","education","everything","hack","hacking","llama","llamacpp","llm","llms","python","software","uncensored","uncensored-llm","uncensoredai","unleashedllm"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/0x0806.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-07-21T10:32:10.000Z","updated_at":"2025-07-22T13:35:55.000Z","dependencies_parsed_at":"2025-07-22T14:39:53.008Z","dependency_job_id":"509f5479-85e0-4443-9c83-277efd7ad2eb","html_url":"https://github.com/0x0806/UnleashedLLM","commit_stats":null,"previous_names":["0x0806/unleashedllm"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/0x0806/UnleashedLLM","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0x0806%2FUnleashedLLM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0x0806%2FUnleashedLLM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0x0806%2FUnleashedLLM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0x0806%2FUnleashedLLM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/0x0806","download_url":"https://codeload.github.com/0x0806/UnleashedLLM/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0x0806%2FUnleashedLLM/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":267951759,"owners_count":24171096,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-07-30T02:00:09.044Z","response_time":70,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","aimodels","download","education","everything","hack","hacking","llama","llamacpp","llm","llms","python","software","uncensored","uncensored-llm","uncensoredai","unleashedllm"],"created_at":"2025-07-30T22:16:05.439Z","updated_at":"2026-04-19T17:02:08.000Z","avatar_url":"https://github.com/0x0806.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# UnleashedLLM\n\n**Democratizing Access to Uncensored AI Language Models**\n\nA sophisticated command-line interface tool for downloading, managing, and interacting with powerful AI models completely offline.\n\n---\n\n## Table of Contents\n\n- [Overview](#overview)\n- [Key Features](#key-features)\n- [Supported Models](#supported-models)\n- [System Requirements](#system-requirements)\n- [Installation](#installation)\n- [Usage](#usage)\n- [Architecture](#architecture)\n- [Configuration](#configuration)\n- [API Reference](#api-reference)\n- [Performance Optimization](#performance-optimization)\n- [Troubleshooting](#troubleshooting)\n- [Development](#development)\n- [Security \u0026 Privacy](#security--privacy)\n- [Contributing](#contributing)\n- [Changelog](#changelog)\n- [License](#license)\n- [Support](#support)\n\n---\n\n## Overview\n\nUnleashedLLM is a comprehensive Python application that provides seamless access to uncensored AI language models through an intuitive command-line interface. The tool enables users to download, manage, and interact with various AI models while ensuring complete privacy and offline functionality after initial setup.\n\n### Mission Statement\n\nTo democratize access to powerful AI language models by providing a user-friendly, privacy-focused platform that operates entirely offline without compromising on functionality or performance.\n\n---\n\n## Key Features\n\n### Core Functionality\n- **Advanced Model Management**: Download, organize, and manage AI models from a curated registry\n- **Interactive Chat Interface**: Engage in real-time conversations with context-aware AI models\n- **Single Inference Engine**: Perform one-off text generation tasks with optimized parameters\n- **Complete Offline Operation**: Full functionality without internet dependency post-installation\n\n### Advanced Capabilities\n- **Paginated Model Browser**: Browse available models with detailed specifications and filtering\n- **Real-time Progress Tracking**: Download progress with speed indicators and ETA calculations\n- **Comprehensive System Diagnostics**: Automated system requirement validation and performance analysis\n- **Status Dashboard**: Monitor storage usage, model availability, and system health\n- **Conversation History Management**: Maintain context across extended chat sessions\n- **Thread-safe Operations**: Concurrent model operations with robust error handling\n\n### Technical Features\n- **GGUF Format Support**: Optimized for CPU inference with quantized models\n- **Dynamic Memory Management**: Adaptive memory allocation based on model requirements\n- **Configurable Threading**: Optimized CPU utilization with thread limiting\n- **Automatic Dependency Resolution**: Seamless installation of required packages\n\n---\n\n## Supported Models\n\n### Model Categories\n\n| Category | Size Range | Memory Requirement | Performance Level |\n|----------|------------|-------------------|-------------------|\n| Lightweight | 1-3GB | 4GB RAM | Fast inference, basic capabilities |\n| Medium | 4-8GB | 8GB RAM | Balanced performance and quality |\n| Large | 7-15GB | 16GB RAM | High quality, advanced reasoning |\n| Extra Large | 20GB+ | 32GB RAM | Maximum capability, research-grade |\n\n### Available Models\n\n| Model | Size | Context Length | Capabilities | Specialization |\n|-------|------|----------------|--------------|----------------|\n| Phi-2 Uncensored | 1.6GB | 2,048 tokens | Text generation, Chat | Lightweight conversations |\n| Llama2 7B Uncensored | 4.1GB | 4,096 tokens | Text generation, Chat, Coding | General purpose |\n| OpenChat 7B | 4.2GB | 8,192 tokens | Text generation, Chat, Reasoning | Conversational AI |\n| CodeLlama 13B | 7.3GB | 16,384 tokens | Coding, Text generation | Code generation |\n| Mixtral 8x7B | 26.9GB | 32,768 tokens | Text generation, Chat, Reasoning | Mixture of experts |\n| Llama2 70B Uncensored | 39.5GB | 4,096 tokens | Advanced reasoning, Complex tasks | Research applications |\n\n---\n\n## System Requirements\n\n### Minimum Requirements\n- **Operating System**: Linux (Nix-based environment)\n- **Python Version**: 3.11 or higher\n- **Memory**: 8GB RAM\n- **Storage**: 10GB available space (plus model storage)\n- **CPU**: Multi-core processor (4+ cores recommended)\n\n### Recommended Configuration\n- **Memory**: 16GB+ RAM for optimal performance\n- **Storage**: SSD with 100GB+ available space\n- **CPU**: 8+ core processor with high clock speed\n- **Network**: Stable internet connection for initial model downloads\n\n### Performance Benchmarks\n- **Lightweight Models**: 2-5 tokens/second on 4-core CPU\n- **Medium Models**: 1-3 tokens/second on 8-core CPU\n- **Large Models**: 0.5-1.5 tokens/second on high-end CPU\n\n---\n\n## Installation\n\n### Quick Start\n\n```bash\n# Clone the repository\ngit clone \u003crepository-url\u003e\ncd UnleashedLLM\n\n# Run the application (dependencies auto-install)\npython3 main.py\n```\n\n### Manual Dependency Installation\n\n```bash\n# Install core dependencies\npip install llama-cpp-python\n\n# Verify installation\npython3 -c \"import llama_cpp; print('Installation successful')\"\n```\n\n### Environment Setup\n\n```bash\n# Set environment variables (optional)\nexport LLAMA_CPP_LOG_LEVEL=ERROR\nexport OMP_NUM_THREADS=8\n```\n\n---\n\n## Usage\n\n### Command-Line Interface\n\nLaunch the main application:\n\n```bash\npython3 main.py\n```\n\n### Menu Navigation\n\nThe application provides an intuitive menu system:\n\n1. **Browse Model Library** - Explore available models with detailed specifications\n2. **Download Model** - Select and download models with progress tracking\n3. **Interactive Chat** - Start conversational sessions with downloaded models\n4. **Single Inference** - Perform individual text generation tasks\n5. **Manage Models** - View, organize, and delete downloaded models\n6. **System Diagnostics** - Check system compatibility and performance metrics\n7. **Status Dashboard** - Monitor application status and resource usage\n8. **Exit** - Safely terminate the application\n\n### Advanced Usage Examples\n\n#### Batch Model Download\n```bash\n# Download multiple models in sequence\npython3 main.py --batch-download llama2-7b-uncensored,openchat-7b\n```\n\n#### Configuration Override\n```bash\n# Override default parameters\npython3 main.py --max-threads 4 --context-length 2048\n```\n\n#### Automated Chat Session\n```bash\n# Start chat with specific model\npython3 main.py --chat --model llama2-7b-uncensored\n```\n\n---\n\n## Architecture\n\n### Core Components\n\n#### ModelRegistry\n- Centralized model metadata management\n- Version control and compatibility tracking\n- Dynamic model discovery and registration\n\n#### ModelManager\n- File system operations and storage management\n- Download resumption and integrity verification\n- Model lifecycle management\n\n#### ChatInterface\n- Context-aware conversation handling\n- Session persistence and history management\n- Response formatting and sanitization\n\n#### LlamaCppManager\n- Direct integration with llama-cpp-python\n- Memory optimization and thread management\n- Performance monitoring and adjustment\n\n#### SystemDiagnostics\n- Real-time system monitoring\n- Performance benchmarking\n- Resource usage tracking\n\n### Design Patterns\n\n- **Singleton Pattern**: Ensures single instance of critical managers\n- **Factory Pattern**: Dynamic model instantiation based on configuration\n- **Observer Pattern**: Event-driven status updates and notifications\n- **Strategy Pattern**: Pluggable inference backends and optimization strategies\n\n### Data Flow\n\n```\nUser Input → CLI Parser → Model Manager → Inference Engine → Response Formatter → User Output\n     ↓              ↓              ↓              ↓                    ↑\nSystem Check → Model Registry → Model Loading → Context Management → Result Processing\n```\n\n---\n\n## Configuration\n\n### Default Parameters\n\n```python\nDEFAULT_CONFIG = {\n    \"max_threads\": 8,\n    \"context_length\": \"auto\",  # Dynamically set per model\n    \"batch_size\": 512,\n    \"temperature\": 0.8,\n    \"top_p\": 0.95,\n    \"top_k\": 40,\n    \"repeat_penalty\": 1.1,\n    \"max_tokens\": 400\n}\n```\n\n### Environment Variables\n\n| Variable | Description | Default |\n|----------|-------------|---------|\n| `LLAMA_CPP_LOG_LEVEL` | Logging verbosity | `ERROR` |\n| `OMP_NUM_THREADS` | OpenMP thread count | `8` |\n| `MODEL_CACHE_DIR` | Model storage directory | `./models` |\n| `MAX_CONTEXT_LENGTH` | Maximum context tokens | `32768` |\n\n### Storage Structure\n\n```\nUnleashedLLM/\n├── main.py                 # Main application entry point\n├── models/                 # Downloaded model storage\n│   ├── llama-2-7b-chat.Q4_K_M.gguf\n│   ├── openchat-3.5-0106.Q4_K_M.gguf\n│   └── ...\n├── cache/                  # Runtime cache and temporary files\n├── logs/                   # Application logs and diagnostics\n├── config/                 # User configuration files\n└── README.md              # Documentation\n```\n\n---\n\n## API Reference\n\n### Core Classes\n\n#### ModelRegistry\n```python\nclass ModelRegistry:\n    @staticmethod\n    def get_model_info(model_id: str) -\u003e Dict\n    @staticmethod\n    def list_models_by_category(category: str) -\u003e List[Dict]\n    @staticmethod\n    def validate_model(model_data: Dict) -\u003e bool\n```\n\n#### ModelManager\n```python\nclass ModelManager:\n    def download_model(self, model_id: str, progress_callback=None) -\u003e bool\n    def list_downloaded_models(self) -\u003e List[str]\n    def delete_model(self, model_id: str) -\u003e bool\n    def get_model_path(self, model_id: str) -\u003e Path\n```\n\n#### ChatInterface\n```python\nclass ChatInterface:\n    def start_chat(self, model_id: str) -\u003e None\n    def send_message(self, message: str) -\u003e str\n    def end_chat(self) -\u003e None\n    def get_chat_history(self) -\u003e List[Dict]\n```\n\n---\n\n## Performance Optimization\n\n### CPU Optimization\n- Thread count limited to prevent CPU overload\n- Automatic CPU core detection and utilization\n- Dynamic batch size adjustment based on system performance\n\n### Memory Management\n- Lazy model loading to minimize memory footprint\n- Automatic garbage collection for conversation history\n- Memory-mapped file access for large models\n\n### I/O Optimization\n- Streaming downloads with resume capability\n- Asynchronous file operations\n- Compressed model storage when possible\n\n### Benchmarking Results\n\n| Model Size | RAM Usage | CPU Utilization | Tokens/Second |\n|------------|-----------|-----------------|---------------|\n| 1.6GB | 2.1GB | 45% | 4.2 |\n| 4.1GB | 5.8GB | 65% | 2.8 |\n| 7.3GB | 9.2GB | 78% | 1.9 |\n| 26.9GB | 31.4GB | 85% | 0.8 |\n\n---\n\n## Troubleshooting\n\n### Common Issues\n\n#### Installation Problems\n**Issue**: `llama-cpp-python` compilation fails\n```bash\n# Solution: Install build dependencies\nsudo apt update\nsudo apt install build-essential cmake\npip install --upgrade pip setuptools wheel\n```\n\n#### Memory Errors\n**Issue**: Out of memory when loading large models\n- Reduce context length in configuration\n- Close other memory-intensive applications\n- Consider using smaller model variants\n\n#### Performance Issues\n**Issue**: Slow inference speed\n- Adjust thread count based on CPU cores\n- Ensure sufficient RAM is available\n- Check for background processes consuming resources\n\n#### Download Failures\n**Issue**: Network timeouts during model download\n- Check internet connection stability\n- Verify sufficient disk space\n- Resume download using the same model selection\n\n### Diagnostic Commands\n\n```bash\n# Check system resources\npython3 main.py --diagnostics\n\n# Verify model integrity\npython3 main.py --verify-models\n\n# Performance benchmark\npython3 main.py --benchmark\n```\n\n### Log Analysis\n\nApplication logs are stored in the `logs/` directory:\n- `application.log`: General application events\n- `performance.log`: Performance metrics and benchmarks\n- `errors.log`: Error messages and stack traces\n\n---\n\n## Development\n\n### Project Structure\n\n```\nsrc/\n├── core/\n│   ├── model_registry.py\n│   ├── model_manager.py\n│   ├── chat_interface.py\n│   └── diagnostics.py\n├── utils/\n│   ├── file_operations.py\n│   ├── network_utils.py\n│   └── formatting.py\n├── config/\n│   ├── default_config.py\n│   └── model_definitions.py\n└── tests/\n    ├── unit/\n    ├── integration/\n    └── performance/\n```\n\n### Development Setup\n\n```bash\n# Create development environment\npython3 -m venv dev-env\nsource dev-env/bin/activate\n\n# Install development dependencies\npip install pytest black flake8 mypy\n\n# Run tests\npytest tests/\n\n# Code formatting\nblack src/\nflake8 src/\n```\n\n### Contributing Guidelines\n\n1. **Code Style**: Follow PEP 8 guidelines with Black formatting\n2. **Testing**: Maintain \u003e90% test coverage for new features\n3. **Documentation**: Update README and inline documentation\n4. **Performance**: Benchmark new features for regression testing\n5. **Security**: Review code for potential vulnerabilities\n\n### Extension Points\n\n#### Custom Model Backends\n```python\nclass CustomInferenceEngine:\n    def load_model(self, model_path: str) -\u003e bool\n    def generate(self, prompt: str, **kwargs) -\u003e str\n    def unload_model(self) -\u003e None\n```\n\n#### Plugin Architecture\n```python\nclass UnleashedLLMPlugin:\n    def initialize(self, app_context: Dict) -\u003e None\n    def process_input(self, user_input: str) -\u003e str\n    def cleanup(self) -\u003e None\n```\n\n---\n\n## Security \u0026 Privacy\n\n### Privacy Features\n- **Complete Offline Operation**: No data transmission after model download\n- **Local Storage**: All models and conversations stored locally\n- **No Telemetry**: Zero usage data collection or external communications\n- **Secure Model Verification**: Checksum validation for downloaded files\n\n### Security Considerations\n- **Input Sanitization**: All user inputs are sanitized before processing\n- **File System Isolation**: Models stored in dedicated directories with restricted permissions\n- **Memory Protection**: Sensitive data cleared from memory after use\n- **Audit Trail**: Comprehensive logging of all system operations\n\n### Best Practices\n- Regularly update the application for security patches\n- Monitor system resources for unusual activity\n- Validate model sources before downloading\n- Use appropriate file permissions for model storage\n\n---\n\n## Contributing\n\n### How to Contribute\n\n1. **Fork the Repository**\n   ```bash\n   git clone \u003cyour-fork-url\u003e\n   cd UnleashedLLM\n   ```\n\n2. **Create Feature Branch**\n   ```bash\n   git checkout -b feature/your-feature-name\n   ```\n\n3. **Make Changes**\n   - Follow coding standards\n   - Add comprehensive tests\n   - Update documentation\n\n4. **Submit Pull Request**\n   - Provide clear description of changes\n   - Reference related issues\n   - Ensure CI passes\n\n### Development Workflow\n\n- **Issues**: Use GitHub issues for bug reports and feature requests\n- **Code Review**: All changes require peer review\n- **Testing**: Automated testing with CI/CD pipeline\n- **Documentation**: Keep documentation in sync with code changes\n\n---\n\n## Changelog\n\n### Version 1.1.0 (2025)\n- Enhanced model registry with 30+ models\n- Improved download resumption capabilities\n- Added system diagnostics and performance monitoring\n- Optimized memory management for large models\n- Enhanced chat interface with better context handling\n\n### Version 1.0.0 (2024)\n- Initial release with core functionality\n- Basic model download and chat capabilities\n- Command-line interface implementation\n- Support for GGUF model format\n\n---\n\n## License\n\nThis project is developed for educational and research purposes. Users are responsible for compliance with applicable AI model licenses and terms of use.\n\n### Third-Party Licenses\n- **llama-cpp-python**: MIT License\n- **Model Files**: Various licenses (see individual model documentation)\n\n---\n\n## Support\n\n### Getting Help\n\n- **Documentation**: Comprehensive README and inline documentation\n- **Issues**: GitHub Issues for bug reports and feature requests\n- **Community**: Discussion forums and community support\n\n### Contact Information\n\n- **Developer**: 0x0806\n- **Project Repository**: [GitHub Repository URL]\n- **Issue Tracker**: [GitHub Issues URL]\n\n---\n\n**UnleashedLLM v1.1.0** | **Python 3.11+** | **Linux/Nix Environment**\n\n*Last Updated: 2025*\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0x0806%2Funleashedllm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F0x0806%2Funleashedllm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0x0806%2Funleashedllm/lists"}