HKU simulates 12 million households on 14 GPUs in under 4 hours
机器之心 · wechat · 2026-11-06
The University of Hong Kong's S²-MAS framework scaled LLM-driven social simulation to 12 million synthetic households using just 14 GPUs in under 4 hours. It combines a 500M-persona database, YN-mini — a 421M-parameter scoring-head model that beats GLM-5.3-Flash and GPT-6-Astra on 240K consumer-finance questions while hitting 37K tokens/s on one A100 — and the Genese engine (2.12-4.10x faster scheduling than peers). The run covered 24M paired decisions and 288M monthly ledger states. Models and code are open-sourced.
More from coding & agent
- Gemini /teamwork agents caught live: four parallel task logs all running that second — RyanSpunk · 2026-11-13
- Codex CLI 0.168.0 adds model provider picker, side-by-side TUI workspace and animated GIF generation — github-actions[bot] · 2026-11-13
- Free MCP server lets Claude read Altium, Allegro, KiCad and Gerber PCB files directly — woodisruff · 2026-11-13
- a16z Partner Clones Open-Source Game with Fable 5.2 Agent, Which Finds Its Own Horse Animation Asset — FinanceYF5 · 2026-11-13
- OpenAI team shares how it runs and governs fleets of persistent agents internally — romainhuet · 2026-11-13
- Perplexity's Decider scanned 1B tokens across 7,000 agent sessions, found 14 security fixes — denisyarats · 2026-11-13