diogovasconcelosmerca/multi-agent-debate: Compare single-agent LLM responses against a propose/critique/revise/judge multi-agent debate. Four interchangeable backends (Ollama, Groq, Gemini, OpenRouter), Pydantic-validated outputs, structured telemetry, golden-dataset evals.

Source-linked topic cluster with 1 signals across related articles, projects, models, papers, and source updates.