Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Introducing System One Models and Jev(typesafe.ai ↗)
    395comments
  2. Apple Reference Image: A New Approach for Verified Photography(security.apple.com ↗)
    168comments
  3. Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations(github.com/arnegiacomo ↗)
    205comments
  4. Show HN: I made a flight simulator, except you're just a passenger(inflightsimulator.com ↗)
    95comments
  5. Learning Programming in an Age of LLMs(ploeh.dk ↗)
    discuss
  6. An update on Wayback Machine access(blog.archive.org ↗)
    271comments
  7. Doing Everyone Else's Job(yosefk.com ↗)
    32comments
  8. Gemini 3.8 Live and 3.8 Live Extended Thinking(blog.google ↗)
    269comments
  9. Negativland, Culture Jamming, and the Art of Making Something New(blog.archive.org ↗)
    19comments
  10. Building a Linux GPU Driver for the M4 Mac Mini in One Month(codyho.dev ↗)
    182comments
  11. German Rheinmetall open-sources its Battlesuite connected weapon system protcol(rheinmetall.github.io ↗)
    77comments
  12. Recreating Voodoo Graphics and a Late-1990s Gaming PC on an FPGA(nand2mario.github.io ↗)
    37comments
  13. Why I'm still bearish on LLMs after Navier-Stokes(dank.systems ↗)
    248comments
  14. A software thing I built: GPS on a 25MHz 486-SX(vcfed.org ↗)
    6comments
  15. Intelligence per Watt: Measuring Intelligence Efficiency of Local AI(arxiv.org ↗)
    3comments
  16. Datamimic – don't let your coding agent invent its own test world(github.com/rapiddweller ↗)
    6comments
  17. We got admin access to Baseten's production GitHub(strix.ai ↗)
    153comments
  18. MartyPC – A Cycle-Accurate IBM PC/XT Emulator(github.com/dbalsom ↗)
    1comments
  19. Saving Jet Fuel(marksblogg.com ↗)
    44comments
  20. Show HN: Capsule – Single-file web apps that save their data into SQLite(withcapsule.app ↗)
    139comments
  21. Let's make quality the norm again(forbrukerradet.no ↗)
    394comments
  22. Learning to solve hard problems in RL for LLMs by never giving up(mnoukhov.github.io ↗)
    4comments
  23. Mistral X Mozilla: Private, Multilingual AI Browsing(mistral.ai ↗)
    discuss
  24. An interactive world map of the stories cultures have told(sunnyguha.com ↗)
    6comments
  25. Chopping up books when they're physically too big(mattkirkland.com ↗)
    165comments
  26. Suspected sabotage causes major Netherlands rail disruption(bbc.com ↗)
    427comments
  27. Jean-Pierre Serre turns 100(st-andrews.ac.uk ↗)
    24comments
  28. The Beauty of Roundabouts(gruhn.me ↗)
    103comments
  29. WangNet – 1.8 MB, zero-dependency Numberwang adjudication in 11 languages(github.com/graafhenk ↗)
    56comments
  30. Stay discoverable in search while disallowing AI training(cloudflare.com ↗)
    41comments

Intelligence per Watt: Measuring Intelligence Efficiency of Local AI

37 pointsby 2d agoarxiv.org
3 comments
1h agoHN ↗

We propose intelligence per watt (IPW), task accuracy per unit of power

Stupid metric. It's not because a model is better performing that it necessarily requires more energy or compute.

55m agoHN ↗

I don't think that's what they are saying. In fact if they did the metric would be pointless. Rather they are saying by estimating that value on different architectures, one can find more efficient ones. They use open model to be able to remove unknowns. They aren't advocating for one model or another, only more efficient architectures.

32m agoHN ↗

We propose miles per hour (MPH), distance travelled per unit of time

Stupid metric. It‘s not because you spend more time that you travel farther.

s/