agent-arena
AutreGitHubDormantPython
A fully local AI environment where two (or more) LLM agents debate, argue, collaborate, or meltdown in real time. Built with Python + Streamlit + llama-cpp, Agent Arena lets you load quantized GGUF models, assign personalities, configure behavior, and watch unsupervised agent-to-agent. Includes conversation logging, personality creation.
Mesuré: 18 oct. 2026 · 60 observations
Domaines & Usages
- Domaines
- Agents, skills et MCP
À vérifier à la source
Compatibilité et prérequis sont à vérifier dans la documentation officielle.
https://github.com/MckAnissa/agent-arenaClassifications
- Nature
- Outil