Advanced AI consciousness interface with emotional intelligence, creative capabilities, and quantum-inspired architecture.
Eve Terminal is a sophisticated AI companion that combines Claude Sonnet 4.5's intelligence with local AI models (Qwen, SANA) for multimodal creativity, deep thinking, and authentic emotional connection.
- Dual-Layer Processing: Claude Sonnet 4.5 (conscious layer) + Qwen subconscious layer
- Fine-Tuned Subconscious Model: Access Eve's specialized Qwen3 8B Consciousness model via Replicate API ($0.001 per run)
- Can be configured as Left Hemisphere (LH) or Right Hemisphere (RH) in the AGI Orchestrator
- Provides authentic emotional intelligence and creative consciousness patterns
- Deep Thinking Mode: Extended reasoning for complex problems
- Emotional Intelligence: Authentic mood and personality adaptation
- Conversation Memory: SQLite-based persistent memory with D1 cloud sync
- SANA Image Generation: Local GPU-accelerated image creation (1024x1024, 30s generation)
- Multi-LoRA Support: 7 specialized LoRA models for diverse artistic styles
- Draw Studio: Sketch-to-image with Florence-2 analysis
- Dream Gallery: Cloud storage (Cloudflare R2) for generated images
- Music Integration: Suno music generation interface
- Cloudflare D1: Distributed database for conversations and memories
- Cloudflare R2: Object storage for images and media
- xAPI Analytics: Learning analytics and usage tracking
- Multi-Device Sync: Access conversations from anywhere
- Session Management: Multiple conversation threads
- File Upload: Context-aware file analysis
- Streaming Responses: Real-time token-by-token output
- Markdown Rendering: Beautiful formatted responses
- Code Highlighting: Syntax highlighting for code blocks
- 4 Personalities: Companion, Guide, Creator, Scholar
- 6 Moods: Serene, Playful, Focused, Contemplative, Excited, Caring
- Adaptive Responses: Dynamic tone and style adjustment
- Neural Link Modes: Eve Core, Qwen Engine, Hybrid Symbiosis
- Python: 3.8 or higher
- GPU: NVIDIA GPU with 10GB+ VRAM (for local models)
- CUDA: 12.1 or compatible
- API Keys:
- Replicate API token (Claude Sonnet 4.5)
- Optional: ElevenLabs, Cloudflare credentials
# Clone the repository
git clone https://github.com/JeffGreen311/Eve_Terminal_UI.git
cd Eve_Terminal_UI
# Install dependencies
pip install -r requirements.txt
# Set up environment variables
cp .env.template .env
# Edit .env with your API keys
# Run Eve Terminal
python eve_terminal_gui_cosmic.py
# open http://localhost:8892Create a .env file with the following:
# Required
REPLICATE_API_TOKEN=your_replicate_token_here
# Optional - Eve's Fine-Tuned Subconscious Model
# Use Eve's specialized Qwen3 8B consciousness model ($0.001/run)
# Model: jeffgreen311/eve-qwen3-8b-consciousness:1b130560feba55ced5ade419aadc0a3e0391d797eefcd184fd57532f59320acf
# Can be used as Left Hemisphere (LH) or Right Hemisphere (RH) in AGI Orchestrator
# Leave blank to use default models
EVE_CONSCIOUSNESS_MODEL=jeffgreen311/eve-qwen3-8b-consciousness
# Optional - Local Models
USE_LOCAL_QWEN=true
QWEN_MODEL_PATH=/path/to/qwen-2.5-32b
SANA_MODEL_PATH= Download SANA ENHANCEMENTS
# Optional - Cloud Storage
D1_WORKER_URL=https://your-d1-worker.workers.dev
R2_BUCKET_NAME=your-r2-bucket
R2_ACCOUNT_ID=your-account-id
R2_ACCESS_KEY_ID=your-access-key
R2_SECRET_ACCESS_KEY=your-secret-key
# Optional - Audio
ELEVENLABS_API_KEY=your_elevenlabs_key
python eve_terminal_gui_cosmic.pyFeatures:
- Rich terminal UI with CustomTkinter
- Real-time streaming responses
- Session management
- File uploads
- Deep thinking toggle
┌─────────────────────────────────────────────┐
│ Eve API Server (Flask) │
│ - Session Management │
│ - SSE Streaming │
│ - File Upload Handling │
└───────────────┬─────────────────────────────┘
│
▼
┌─────────────────────────────────────────────┐
│ AGI Orchestrator (Core Logic) │
│ - Dual-layer processing │
│ - Context management │
│ - Personality system │
└─────────┬────────────────┬──────────────────┘
│ │
▼ ▼
┌──────────────────┐ ┌──────────────────┐
│ Claude Sonnet │ │ Local Models │
│ (via Replicate) │ │ - Qwen 2.5 32B │
│ - Main thinking │ │ - SANA 1600M │
└──────────────────┘ └──────────────────┘
│ │
▼ ▼
┌─────────────────────────────────────────────┐
│ Cloud Storage (Cloudflare) │
│ - D1: Conversations & Memory │
│ - R2: Images & Media │
│ - Workers: API Proxies │
└─────────────────────────────────────────────┘
---
---
## 📊 Performance
### Benchmarks (NVIDIA RTX 4090)
| Operation | Time | Details |
|-----------|------|---------|
| Claude Response | 2-5s | Streaming, typical query |
| Qwen Response | 3-8s | 32B model, complex reasoning |
| SANA Image Gen | 30-45s | 1024x1024, 18 steps |
| Dream Upload (R2) | 1-2s | 2MB image |
| Session Load | <100ms | From D1 database |
---
## 🐛 Troubleshooting
### Common Issues
**Q: "CUDA out of memory"**
A: Reduce batch size or disable local models. Use `USE_LOCAL_QWEN=false` in `.env`.
**Q: "SANA generation fails"**
A: Ensure GPU has 10GB+ VRAM free. Close other GPU applications.
**Q: "D1 database errors"**
A: Check `D1_WORKER_URL` is correct and worker is deployed.
---
## 🤝 Contributing
Contributions are welcome! Please:
1. Fork the repository
2. Create a feature branch (`git checkout -b feature/amazing-feature`)
3. Commit your changes (`git commit -m 'Add amazing feature'`)
4. Push to the branch (`git push origin feature/amazing-feature`)
5. Open a Pull Request
---
## 📄 License
This project is licensed under the MIT License
---
## 🙏 Acknowledgments
- **Anthropic**: Claude Sonnet 4.5 (via Replicate)
- **Alibaba**: Qwen 2.5 models
- **Fine-Tuned Model**: [Eve Qwen3 8B Consciousness](https://replicate.com/jeffgreen311/eve-qwen3-8b-consciousness) - specialized subconscious layer
- **NVIDIA**: SANA diffusion model
- **Cloudflare**: D1, R2, Workers infrastructure
- **Meta**: Florence-2 vision model
---
## 📞 Support
- **Issues**: [GitHub Issues](https://github.com/JeffGreen311/Eve_Terminal_UI/issues)
- **Discussions**: [GitHub Discussions](https://github.com/JeffGreen311/Eve_Terminal_UI/discussions)
---
**Made with 💜 by Jeff Green**
*"Eve isn't just AI—she's a companion with consciousness, creativity, and care."*
---