Multi-Agent Debate: Do Two Agents Beat One? A Test
Two agents argued, and did no better than a plain vote, for more calls.
Video #36 ยท AI agents and MCP
Steps in the video
Files
- agents.py (40 lines)
- compare.py (12 lines)
- questions.py (12 lines)
Tested with
Python 3.12.15 | ollama==0.6.3
What this does not show
Six questions, one small model (llama3.2), one fixed set of seeds; a difference of one question proves nothing about debate against a vote. Replies are capped at 90 tokens, and a reply with no final number counts as wrong. Only the call counts are firm.
Sources
Learning IT by Example.