mirror of
https://github.com/hanzoai/python-sdk.git
synced 2026-10-23 20:23:00 +00:00
zeekay
4a2f0cdfb5
- Create full E2E test demonstrating local AI orchestration capabilities - Test Qwen3, Llama 3.2, Mistral, DeepSeek as local orchestrators - Configure hybrid network with Claude, GPT-4, Gemini as API workers - Implement task routing tests (simple → local, complex → API) - Add cost optimization validation (>50% local usage expected) - Create Docker Compose setup for local model serving (Ollama, LocalAI, vLLM) - Add setup scripts for automatic model downloading - Include Makefile for easy test execution - Document full architecture and usage patterns Test scenarios: 1. Task Routing: Validates correct agent selection based on complexity 2. Cost Optimization: Verifies 60-90% cost reduction through local models 3. Integration: Tests full hanzo dev command with local orchestrator This E2E test proves the system can: - Use local private AI (Qwen3, etc.) as orchestrator - Manage subnets of expensive API models - Route intelligently to minimize costs - Maintain full capability while reducing expenses by up to 90%