Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms (github.com/firelex)
    145comments
  2. 1996 chat room simulator connected to Win95 and System 7 web desktops (lolchat.rip)
    34comments
  3. Pirating the Pirates (mubi.com)
    241comments
  4. MicroLLM Lab – Try 7 tiny LLM's in the browser (stateofutopia.com)
    68comments
  5. 12,000-year-old Göbeklitepe burials explain scattered bones (archaeologymag.com)
    27comments
  6. Tank Body Problem (jimsitu.com)
    9comments
  7. California farmers are struggling to sell grapes as demand for wine drops (kqed.org)
    265comments
  8. ESP32S3 cluster running 1.58-bit (BitNet) Language model (github.com/low-zi-hong)
    8comments
  9. Phyllotaxis: An audio-reactive LED display (jagi.studio)
    1comments
  10. Sonnet 5.5 (anthropic.com)
    447comments
  11. Scientists solve 1840s space weather mystery (arstechnica.com)
    41comments
  12. Who Killed Paulina Borsook's Career? (wired.com)
    —discuss
  13. Show HN: Pac-Bench – How well can models one-shot a Pac-Man game? (jonclegg.github.io)
    5comments
  14. Hijacking the PS5's RTMP stream (yashgarg.dev)
    69comments
  15. World Labs Is Joining AMD (worldlabs.ai)
    93comments
  16. Kids turned low-traffic NPR Spotify comments into a secret group chat (thisamericanlife.org)
    194comments
  17. Does Reddit have an astroturfing problem? What the data suggests (petervijeh.com)
    183comments
  18. How to win a beer with high-dimensional statistics (jamiesimon.io)
    6comments
  19. Updated Google Maps shows destruction of the city of Rafah (twitter.com/aliabunimah)
    172comments
  20. Nvidia wants to put a watchdog chip next to every AI agent (cnbc.com)
    161comments
  21. What is the best shape of a city? Modelling effect of urban form on distance (sagepub.com)
    12comments
  22. Bluegraph – Explore NOAA buoy data, rebuilt in 3D from measured spectra (bluegraph.io)
    2comments
  23. It's Time to Investigate the AI Labs (calnewport.com)
    135comments
  24. The Art Forger Who Became a National Hero (priceonomics.com)
    5comments
  25. U.S. Strategic Petroleum Reserve Falls to Lowest Level Since 1982 (oilprice.com)
    122comments
  26. Show HN: HN.watch – Videos of all Hacker News posts (hn.watch)
    86comments
  27. Cf: The Agentic CLI for the Cloudflare API (cloudflare.com)
    59comments
  28. Profit Margins of the Largest Companies (visualcapitalist.com)
    3comments
  29. What reversing, modernising old games tells us about the economic impact of AI (isfine.org)
    39comments
  30. Behold the pawpaw (cbc.ca)
    26comments

Show HN: Pac-Bench – How well can models one-shot a Pac-Man game?

11 pointsby 5h agojonclegg.github.io
5 comments
Benchmarks how well Harness+models can create a Pac-Man game from a single prompt:

“Create a Pac-Man game in a single HTML page”

Each model gets one shot — no follow-up prompts or fixes.

5h agoHN ↗

Opus 5-5 seemed like a perfect clone, with others displaying flaws in an initial look. Astra notably created a bunch of surrounding ugly crap to look at.

5h agoHN ↗

I was just coming here to say this. I looked at all of them, and Opus 5.5 at high effort seems significantly better than every other clone.

Buffered controls, different pathfinding AI for each ghost, level transitions, etc.

Gpt-5.6 sol high technically completed the assignment as well, but the gaps within the pipes used to construct the maze, the lack of pause when eating a ghost, etc made it feel far less polished.

3h agoHN ↗

I tried this with DOOM. Fable 5 did a pretty shit job. Astra made pretty crazy animated sprites and was pretty good considering.

The fact that these are at all playable and 100x my programming skill level is pretty depressing from a certain pov. The ThreeJS dude posted ab how demotivated he was to continue his work, and while I was never a dev that did much with webgl, I commiserate.

22m agoHN ↗

I wonder if we need better programming abstractions/ languages that can make programming easier for people.

It seems like a lot of the programming is cookie cutter type stuff (where these ai programs shine) and should be easier.

They can be amazing, but promoting “make a pac man game” seems like an exercise in which model has the best training code that was Pac-Man.