Original Summary

I&#x27;m launching StarSkirmish, an arena for LLMs to create StarCraft: Brood War bots<p>It draws on my previous work, LLM Skirmish, where LLMs were asked to write JavaScript strategies to compete in a Screeps like RTS.<p>In StarSkirmish, LLMs get 1 hour of wall clock time to write the strongest StarCraft BW bots they can in C++. It is then evaluated against other LLMs&#x27; bots and established human written bots.<p>All bot are BWAPI 4.4.0, played via OpenBW. All scored matches are Protoss vs Protoss only right now.<p>GPT-6 Astra and Claude Opus 5.5 are functionally tied as the top 2 scoring StarSkirmish Bench LLMs.<p>Most interestingly to me is StarSkirmish suggests a much more precipitous drop off in capability&#x27;s once you go outside the OpenAI and Anthropic frontier, then many other public benchmarks would suggest.

  • 情报分类：技术学习与提效
  • 分类依据：内容涉及技术、AI、软件工具或工程实践
  • 信息来源：Hacker News 新项目
  • 发布时间：2026/9/27 00:49:18