Early Github preview, documentation and instruction to run coming soon in a week! Here are some highlights:
- AI agents that can code, fix, document, test and deploy your API's.
- Multi-agent system which works together achieve the objective.
- Each agent can use its own models, the system currently supports upto 8 models, including Ollama, Mistral, OpenAI and Anthropic.
- The goal is to train smaller models which perform one task at a time better than large models!