1,000 agents, one week, 630k lines: HF builds a weights-free Python chatbot
_lewtun · x · 2026-11-16
A Hugging Face experiment had 1,000 DeepSeek-V4.1-Flash agent instances collaboratively build "PyBot" — a chatbot made purely of 1M rules, no weights, no training. The agents coordinated on a Slack-like message board (175k messages) and merged 8k PRs on a shared repo (nearly breaking the Hub). Result: 630k lines of zero-dependency Python, roughly matching SmolLM-360M on everyday Q&A. Code and demo are public.
More from coding & agent
- Self-improving agent papers this week: agents learn the benchmark, not the task — TheTuringPost · 2026-11-16
- One-person content agency targets $30K/mo with 11 AI agents — full playbook — VibeMarketer_ · 2026-11-16
- Hugging Face releases definitive fine-tuning guide for 3D generation model TRELLIS.2 — ariG23498 · 2026-11-16
- WebArena-Pro: a 300-task multimodal benchmark across 20 self-hosted apps for web agents — McGill-NLP · 2026-11-16
- Continual-ARC: lifelong ARC-AGI stream finds agent harness persistence rarely beats simple caching — unifyai · 2026-11-16
- COrigami open-sourced: point any coding agent at it to go from prompt to foldable crease pattern — TZahavy · 2026-11-16