jkbennitt/felix-framework
๐ช๏ธ Felix Framework
Helix-Based Multi-Agent Cognitive Architecture
Felix Framework revolutionizes multi-agent systems by replacing traditional graph-based orchestration with 3D helix-based cognitive architecture. Instead of explicit state machines, agents naturally converge through geometric spiral paths, creating emergent coordination patterns.
โก Live Demo Features
This interactive demo showcases the Felix Framework's unique approach to multi-agent coordination:
- ๐ช๏ธ Helix-Based Architecture: Agents spiral from broad exploration to focused synthesis
- โก ZeroGPU Acceleration: GPU-optimized processing for real-time agent coordination
- ๐ Real-time Visualization: 3D interactive helix with agent position tracking
- ๐ฏ Multiple Agent Types: Research, Analysis, Synthesis, and Critic agents
- ๐ฑ Mobile Responsive: Works seamlessly on all devices
- ๐ฌ Research Validated: Statistically significant performance improvements
๐ฎ How to Use
1. Interactive Demo Tab
- Enter a topic you want explored by the multi-agent system
- Select agent types (Research, Analysis, Synthesis, Critic)
- Choose complexity level (Demo: 3 agents โ Research: 20 agents)
- Watch real-time coordination as agents spiral through the helix
2. 3D Helix Explorer
- Visualize the geometric model underlying the cognitive architecture
- Track agent positions as they move from broad (top) to focused (bottom)
- Interactive controls for camera angles and filtering
3. Performance Dashboard
- Monitor system performance with real-time metrics
- GPU utilization tracking with ZeroGPU optimization
- Compare architectures (Felix vs LangGraph vs Mesh)
4. Educational Content
- Learn the mathematics behind helix-based coordination
- Research validation results with statistical significance
- Framework comparisons with traditional approaches
๐ฌ Research Foundation
Felix Framework is built on rigorous research with validated mathematical models:
- Mathematical Precision: <1e-12 error tolerance in geometric calculations
- Statistical Validation: 2/3 hypotheses supported with significance (p<0.05)
- Performance Metrics: 75% memory efficiency improvement over mesh architectures
- Publication Ready: Research-grade methodology and documentation
Key Research Results
- H1 SUPPORTED: Helix shows better task distribution efficiency (p=0.0441)
- Memory Efficiency: O(N) communication vs O(Nยฒ) for mesh architectures
- Processing Speed: Sub-2s coordination time for 20-agent tasks
- Scalability: Linear scaling to 133+ agents demonstrated
๐ ๏ธ Architecture Highlights
Helix Geometry
- 33 spiral turns with geometric tapering from radius 33 to 0.001
- Natural attention focusing through 4,119x concentration ratio
- Position-aware processing with temperature adjustment based on helix position
ZeroGPU Optimization
- @spaces.GPU decorators for compute-intensive operations
- Automatic memory management with intelligent cleanup
- Batch processing for multiple agents on single GPU allocation
- Fallback mechanisms to CPU when GPU unavailable
Agent Coordination
- Spoke-based communication (O(N) complexity) to central coordination system
- Independent spawn timing with natural convergence patterns
- Specialized agent types with unique cognitive functions
- Emergent coordination without explicit state machines
๐ Getting Started Locally
Want to explore the code or contribute? Check out the full repository:
git clone https://github.com/jkbennitt/thefelix.git
cd thefelix
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
# Validate installation
python tests/validation/validate_felix_framework.py
# Run local demos
python examples/blog_writer.py "Your topic here"๐ Documentation
- [HF Spaces Deployment Guide](./docs/hf-spaces/guides/deployment-guide.md) - Deploy your own Felix Space
- [Complete Documentation Hub](./docs/README.md) - Navigation to all documentation
- [Project Summary](./docs/PROJECT_INDEX.md) - Executive overview and status
- [Research Documentation](./RESEARCH_LOG.md) - Complete research journey
- [Mathematical Model](./docs/architecture/core/mathematical_model.md) - Formal geometric foundations
๐ค Contributing
Felix Framework is open-source and welcomes contributions:
- Research Extensions: Explore new hypotheses and validation studies
- Agent Types: Develop specialized agent behaviors
- Visualization: Enhance 3D rendering and interaction
- Performance: Optimize GPU utilization and memory management
๐ License
MIT License - See LICENSE for details.
๐ Credits
Developed with research-grade rigor, validated through comprehensive testing, and optimized for ZeroGPU deployment on HuggingFace Spaces.
Experience the future of multi-agent coordination - where geometry meets artificial intelligence! ๐ช๏ธ
