Minions: where local and cloud LLMs meet
Ollama collaborates with cloud models (e.g., GPT-4o) to shift workloads to local devices, achieving 30.4x cost reduction with 87% performance. MinionS protocol reduces costs 5.7x while maintaining 97.9% performance. Open-source code available.