멀티 에이전트 마켓에서 경쟁 소형 모델 4개 구동 — 마찰의 원인은 없던 CUDA 툴킷뿐이었다
Hugging Face의 'thousand token wood' 시뮬레이션은 gpt-oss-20b, minicpm3-4b, nemotron-mini-4b, qwen 위에서 이기종 에이전트를 구동합니다. 이 실험은 서로 다른 모델 아키텍처가 창발적 경제 행동을 만들어낸다는 점을 증명하는 한편, vllm 설정이 여전히 최대 엔지니어링 병목으로 남아 있음을 보여줍니다.
hugging face released a field report on 'thousand token wood v2', a multi-agent economy where participants operate on distinct small models: gpt-oss-20b (mxfp4), minicpm3-4b, nemotron-mini-4b, and a fine-tuned qwen 0.5b. the project utilized these divergent models—trained on different data with different post-training—to simulate a genuine market where agents 'argue' via trading mechanics rather than following a script. engineering friction was minimal despite the model variety; the major blocker was a vllm dependency issue where the cuda toolkit (nvcc) was missing from lean base images, requiring a switch to a cuda devel image. the project also implemented strict information security protocols, ensuring that secret insider tips provided to agents are stripped from prompts before processing to prevent data leakage, highlighting the architectural requirements for secure multi-agent environments.