Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. You don't need Jev for good emoji search (maxleiter.com)
    —discuss
  2. How Accurate Have AI Progress Forecasts Been So Far? (forecastingresearch.substack.com)
    —discuss
  3. Show HN: Rig, a small operating system for your agent (tryrig.ai)
    —discuss
  4. Playing with the language modeling abilities of gzip (dreamstation.systems)
    —discuss
  5. Talon – hands-free computer control (talonvoice.com)
    —discuss
  6. HomeBody: A Humanoid That Explores, Remembers, and Acts on Its Own (stanford.edu)
    —discuss
  7. Tell-Tale (Sailing) (wikipedia.org)
    —discuss
  8. Bitget blames North Korea for $387.5M crypto wallet raid (theregister.com)
    —discuss
  9. Scientists burned their own scrolls to unlock a 2k-year-old myster (sciencedaily.com)
    —discuss
  10. Coltrane's Shadow (tabletmag.com)
    —discuss
  11. Japan Suddenly Paused All Social Rules [video] (youtube.com)
    —discuss
  12. Making Oneself Obsolete with Slop (techrights.org)
    —discuss
  13. AI can solve math. But can it verify the answer? (lanzar.me)
    —discuss
  14. Cash Is King in San Francisco Real Estate. All-Cash Buyer Market at 20 Year High (sfstandard.com)
    2comments
  15. Nspawn.org: a Docker-like alternative for containers (nspawn.org)
    1comments
  16. Wild Buffalo Make a Comeback on the Canadian Plains (thewalrus.ca)
    —discuss
  17. "I didn't know about this shortcut until recently." (aresluna.org)
    —discuss
  18. Making a Pokémon fan trailer in 2 hours with Opus 5.5 (gist.github.com)
    —discuss
  19. So yeah it was written using AI (berthub.eu)
    —discuss
  20. Can't Publish Link
    —discuss
  21. Don't Worry About AI Making You Dumber, Nvidia CEO Says It Doesn't Matter (thegamer.com)
    1comments
  22. Become Worthless (To Tech Companies) (coryd.dev)
    1comments
  23. Why is website chat still a bubble? (github.com/standchat)
    2comments
  24. FreePascal WebAssembly (WASM) (freepascal.org)
    —discuss
  25. If we do not stop to help each other, what do we become? (codinghorror.com)
    —discuss
  26. Who Is Open Source About? (glyph.im)
    —discuss
  27. Show HN: A short study on where audiophile reviews disagree (audiowords.net)
    1comments
  28. New Europol proposal pushes for deregulation and more AI in policing (statewatch.org)
    —discuss
  29. A Byzantine mosaic you can rewind (echohive.ai)
    —discuss
  30. DeepSeek Elastic Compute (DSec) (arxiv.org)
    —discuss

Show HN: A StarCraft BW Arena where LLMs play by writing code

3 pointsby 2h agostarskirmish.com
1 comments
I'm launching StarSkirmish, an arena for LLMs to create StarCraft: Brood War bots

It draws on my previous work, LLM Skirmish, where LLMs were asked to write JavaScript strategies to compete in a Screeps like RTS.

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' bots and established human written bots.

All bot are BWAPI 4.4.0, played via OpenBW. All scored matches are Protoss vs Protoss only right now.

GPT-6 Astra and Claude Opus 5.5 are functionally tied as the top 2 scoring StarSkirmish Bench LLMs.

Most interestingly to me is StarSkirmish suggests a much more precipitous drop off in capability's once you go outside the OpenAI and Anthropic frontier, then many other public benchmarks would suggest.

1h agoHN ↗

This is awesome, great work thanks for sharing

Games pave the road to ASI?