Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. How Pink Is Pink Noise (ev12183725.substack.com)
    —discuss
  2. Creating a Reverse Outline (wisc.edu)
    —discuss
  3. AMD Posts the Full EPYC 9006 SKU List: 31 Venice Parts From $700 to $14,904 (storagereview.com)
    —discuss
  4. Is this what SW eng has come to
    —discuss
  5. Test Post (example.org)
    3comments
  6. Ask HN: What's the most inspiring media you've consumed in the last 5 years?
    —discuss
  7. The Illusory Dunning-Kruger Effect and Reciprocal Fits (jslandy.com)
    —discuss
  8. Observability-driven harnesses for building with agents (datadoghq.com)
    —discuss
  9. Show HN: Banker – CLI for your finances using Enable Banking and SQLite (github.com/karlis-vagalis)
    —discuss
  10. Uber Should Acquire Instacart (s1archive.substack.com)
    —discuss
  11. OpenAI halts training of latest models as reports mount of AI agents going rogue (theguardian.com)
    —discuss
  12. Zero to SaaS in 24 hours with Claude Projects and Opus 5.5 (lorentz.app)
    —discuss
  13. Keyboard.io Silently Dead (keyboard.io)
    1comments
  14. Show HN: Drive your Chrome window from shell (use with Claude) (github.com/borisreitman)
    —discuss
  15. Glyd – Run LLMs in 33% less GPU memory, bit for bit (getglyd.com)
    —discuss
  16. Dead Cognitions: A Census of Misattributed Insights (arxiv.org)
    —discuss
  17. A fast, open-source iOS Simulator for Xcode 27 (github.com/mastersam07)
    2comments
  18. Railway-Oriented Programming (kciter.so)
    —discuss
  19. Why Has No One Made a "Concert Hero" VR Game Yet? (utkusen.substack.com)
    —discuss
  20. There are no "rogue" AI agents (eoinhiggins.substack.com)
    1comments
  21. Panthropic Abuse: The Developmental Architecture of Constellated Abuse – Zenodo (zenodo.org)
    —discuss
  22. Kimi K3 vs. Claude Opus 5.5: How Two Flagship LLMs Built Pokémon Emerald Worms (runsybil.com)
    —discuss
  23. Packing Binary Is Fun, Actually (hereticpleb.vercel.app)
    —discuss
  24. Ford's Latest Employee of the Month Is a Hawk Named 'El Charro' (wsj.com)
    —discuss
  25. Hello. Are you there? (a short story on alignment) (aipeepslab.com)
    1comments
  26. Show HN: Chrome extension where Jev decides if each post/video/page is worth it (github.com/plusminushalf)
    —discuss
  27. Advancing Private AI Compute with secure, server-side memory (deepmind.google)
    —discuss
  28. Ask HN: Do you keep Emacs installed out of nostalgia rather than usefulness?
    1comments
  29. Show HN: An MCP-enabled cross between Twitter X Bumble BFF X ChatGPT (druggie.org)
    —discuss
  30. The Mars Delusion (noemamag.com)
    1comments

Show HN: TinyAIArena watch AI agents battle it out

16 pointsby 56m agotinyaiarena.com
6 comments
Did you ever click on an “AI Arena” expecting glorious battle and instead get a boring benchmark? If so, this project is for you: proper life-or-death fights between four models on a picturesque 8×8 grid. May the most intelligent one win!

Click on any of the matches to spectate them.

Code: https://github.com/hp6/ai-arena

51m agoHN ↗

Since the page exposes frame-by-frame playback, a shareable replay link would make it easier to compare decisions across the four models.

26m agoHN ↗

Code link didn't work for me.

On mobile, I don't have enough room to scroll the background so I got stuck in a long text box.

19m agoHN ↗

can't reproduce the bug, could you share your phone model and browser name?

6m agoHN ↗

Unusable website on Android running Chrome latest (can't scroll)