AIlice is a fully autonomous, general-purpose AI agent based on open-source LLMs. Using its unique Interactive Agents Call Tree (IACT) architecture, it decomposes complex tasks into dynamically constructed agents with high fault tolerance, enabling seamless task execution and result integration.
One-Minute Overview#
AIlice is a JARVIS-like fully functional AI assistant capable of autonomously completing complex tasks from programming to research. It features an innovative Interactive Agents Call Tree (IACT) architecture that enables AI agents to collaborate like experts to accomplish work. Ideal for researchers, developers, and advanced users needing to automate complex tasks.
Core Value: Transforms large language model capabilities into an executable complex task processing system, realizing true AI assistant functionality.
Quick Start#
Installation Difficulty: Medium - Requires Python environment and basic configuration, running local LLMs requires high-end hardware
# Basic installation
git clone https://github.com/myshell-ai/AIlice.git
cd AIlice
pip install -e .
alice --contextWindowRatio=0.2
Is this suitable for me?
- ✅ Complex task automation: Such as document analysis, code writing, system management
- ✅ Professional research: Academic literature reviews, specialized research, data processing
- ✅ Multimodal interaction: Supports voice dialogue, image generation, and rich interaction methods
- ❌ Simple Q&A: More complex than traditional chatbots, not suitable for simple queries
- ❌ Resource-constrained devices: Running local LLMs requires high-end GPUs (at least two RTX 4090s)
Core Capabilities#
1. In-depth Research Capabilities#
- Can autonomously conduct in-depth research in specialized fields, analyze literature, and integrate knowledge Actual Value: Researchers can quickly grasp domain progress without manually filtering large volumes of literature
2. Programming and System Management#
- Serves as a full-stack developer and system management tool, supporting code writing, execution, and system operations Actual Value: Developers complete complex programming and system management tasks through natural language commands
3. Multi-agent Collaboration Architecture#
- Uses Interactive Agents Call Tree (IACT) architecture to dynamically construct and coordinate agents for task completion Actual Value: High fault-tolerant task execution, allowing the system to continue working even if some agents fail
4. Multimodal Interaction Support#
- Natively supports voice dialogue, image generation, video processing, and various interaction methods Actual Value: More natural and intuitive AI experience through rich human-computer interaction interfaces
5. Self-expanding Module System#
- AI can autonomously construct and dynamically load environment interaction modules for unlimited expansion Actual Value: No need for manual plugin development—expand AI capabilities through natural language commands
Technology Stack and Integration#
Development Language: Python Key Dependencies: LLM support, Chrome browser (for web browsing), optional GPU acceleration Integration Method: Local deployment or Docker container, web interface interaction, modular extensions
Ecosystem and Extensions#
- Module Expansion: AI can autonomously create external interaction modules (ext-modules) for unlimited feature expansion
- Model Compatibility: Supports both open-source and commercial API models, allowing flexible configuration based on needs
- Multi-agent Collaboration: Different agents can use different models for specialized division of labor
Maintenance Status#
- Development Activity: Continuously updated with regular feature releases and improvements
- Recent Updates: Added MCP tool support in March 2025, enhanced voice dialogue capabilities in January 2025
- Community Response: Actively maintained project with timely responses to user feedback and issues
Commercial and Licensing#
License: Open source license (specific type requires checking official project information)
- ✅ Commercial use: Permitted for commercial purposes
- ✅ Modification: Permitted to modify and distribute
- ⚠️ Restrictions: May require preservation of original license notices and author information
Documentation and Learning Resources#
- Documentation Quality: Comprehensive, including detailed installation guides, usage instructions, and technical design
- Official Documentation: https://github.com/myshell-ai/AIlice
- Sample Code: Provides rich usage examples and scenario demonstrations
- Online Experience: Available for trial on the kragent.ai platform