Scalable multi-model AI workflows powered by GitHub Models, Ollama, RAG, CRAG, AI Agents, S/LLMs, and MCP.
Agentic GenAI Orchestration is a modular, extensible framework for building intelligent, multi-agent AI systems. It unifies cloud-hosted and locally-run LLMs under a single orchestration layer, enabling retrieval-augmented generation, corrective reasoning, tool-using agents, and inter-agent communication — all at scale.
Whether you're prototyping a RAG pipeline or deploying a production-grade multi-agent workflow, this repository provides the building blocks to do it efficiently.
| Feature | Description |
|---|---|
| 🌐 GitHub Models | Plug-and-play integration with GitHub's hosted model marketplace |
| 🦙 Ollama | Run open-source LLMs (Llama, Mistral, Phi, etc.) locally with zero cloud cost |
| 📚 RAG | Retrieval-Augmented Generation with vector stores for grounded, factual responses |
| 🔄 CRAG | Corrective RAG — self-evaluating retrieval with fallback web search for higher accuracy |
| 🤝 AI Agents | Tool-using, goal-oriented agents with memory and multi-step reasoning |
| ⚡ S/LLMs | Seamless switching between Small LLMs (efficiency) and Large LLMs (capability) |
| 🔌 MCP | Model Context Protocol for structured, scalable inter-model communication |
| 🔀 Multi-Model Workflows | Orchestrate heterogeneous models across cloud and local runtimes |
- Python 3.10+
- Ollama installed and running locally
- A GitHub Models API token
- (Optional) A vector database — ChromaDB, Qdrant, or FAISS
git clone https://github.com/astro05/agentic-genai-orchestration.git
cd agentic-genai-orchestrationpip install -r requirements.txt- GitHub Models integration
- Ollama local model support
- Basic RAG pipeline
- CRAG with corrective retrieval
- Tool-using AI agents
- MCP server/client implementation
- S/LLM routing
- LangGraph-based multi-agent orchestration
- Streaming support across all model backends
- Agent memory with long-term persistence
- Web UI dashboard for workflow visualization
- Docker / Compose deployment setup
- Benchmarking suite for S/LLM routing strategies
Contributions are welcome! Please open an issue first to discuss what you'd like to change.
- Fork the repository
- Create your feature branch:
git checkout -b feature/my-feature - Commit your changes:
git commit -m 'Add my feature' - Push to the branch:
git push origin feature/my-feature - Open a Pull Request
This project is licensed under the MIT License.
- Ollama — for making local LLMs accessible
- GitHub Models — for democratizing access to frontier models
- LangChain — RAG and agent tooling
- ChromaDB — vector store
- The open-source AI community 🌍
Built with ❤️ for the Agentic AI era