Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. OpenJev(openjev.com ↗)
    20comments
  2. Jemalloc 5.4.0(github.com/jemalloc ↗)
    45comments
  3. The scourge of x86 emulation(fex-emu.com ↗)
    27comments
  4. Astra for Law(openai.com ↗)
    583comments
  5. Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint(prismml.com ↗)
    135comments
  6. Bend – A language that blocks AI mistakes via proof, on CPU and GPU(bend-lang.com ↗)
    223comments
  7. Qwen 3.8 Omni Flash(qwen.ai ↗)
    82comments
  8. Hister: A private search engine for the pages you visit and the files you keep(github.com/asciimoo ↗)
    170comments
  9. Wax motor(wikipedia.org ↗)
    71comments
  10. Pre-Greek: The lost language hidden within Ancient Greek(linguisticdiscovery.com ↗)
    39comments
  11. Replacing Pull Requests with Delta(zed.dev ↗)
    4comments
  12. When the fractional part of a float fixes your shader(crocidb.com ↗)
    1comments
  13. Fujitsu launches made-in-Japan next-generation CPU FUJITSU-MONAKA(global.fujitsu ↗)
    234comments
  14. Microsoft exec called AI scraping 'the largest theft of labor in human history'(techcrunch.com ↗)
    20comments
  15. A heap overflow and SSO misconfiguration to compromise OpenAI internal repos(hacktron.ai ↗)
    155comments
  16. Shapelearn Qwen 3.8 27B (13.1 GB VRAM)(byteshape.com ↗)
    11comments
  17. How to Write with an LLM(sockpuppet.org ↗)
    104comments
  18. Ask A Monk – A digital wilderness for thoughts with no immediate answer(askamonk.online ↗)
    24comments
  19. Flet 1.0 – Build cross-platform apps in Python(flet.dev ↗)
    59comments
  20. Telstra outage: The night a network decided the year was 2006(netnod.se ↗)
    27comments
  21. Speeding up gearhash on ARM64(sam.dev ↗)
    discuss
  22. Diplodocus, Long Thought Exclusively American, Turns Up in Spain(sci.news ↗)
    44comments
  23. Why I didn’t sign the Fields medallists’ letter(gowers.wordpress.com ↗)
    370comments
  24. Apple detectives solved mystery of ancient tree and rewrote the history of fruit(scientificamerican.com ↗)
    12comments
  25. How do we prevent mathemathics from devolving into the Medieval Era of secrecy?(mathoverflow.net ↗)
    108comments
  26. The most important product decision is what you don't build(liamnugent.me ↗)
    39comments
  27. Why Does the Universe Expand?(cosmicave.org ↗)
    58comments
  28. CrowdSec Source Code Leak(crowdsec.net ↗)
    49comments
  29. How Uber Protects Against Retry Storms(uber.com ↗)
    44comments
  30. Show HN: Snapdrop: Instantly share files between devices. No setup, no signup(snapdrop.me ↗)
    39comments

OpenJev

55 pointsby 1h agoopenjev.com
13 comments
42m agoHN ↗

Jev is such a different approach where you have to be specific about what you want and which options are open. Really interesting how those things evolve in usable features for people.

Also with this example the speed of new launches based on a launch is just incredible.

36m agoHN ↗

"... such a different approach where you have to be specific about what you want and which options are open" --- back to where we started ...

30m agoHN ↗

Not sure on that, maybe the options to choose from will be generated and curated. Same as we do with tagging datasets for images. Might be wildly successful for real world decisions.

32m agoHN ↗

This is true Jevons Paradox (hence the Jev name) there will be so many usecases, applications and even new jobs out of this.

Learned also that Jev was trained on 100%(!) synthetic data.

What a great time to be alive.

29m agoHN ↗

Give it a real choice

As opposed to a fake choice?

27m agoHN ↗

I'm confused... This has no relation with the Jev team, isn't it?

It's trying to "emulate" Jev behavior using a regular small LLM model (Qwen3 0.6B or MiniCPM5 2B). And with the smallest model it takes like between half to two seconds to run in my M2 Max, so it's not super fast.

I mean, it's faster than asking to a regular LLM, but I think that's not proper to have Jev on the name (also legally...)

Edit: no shade, and I'll give it a try for some ideas. I'd also like to have an open weights Jev but I think the naming is misguiding. I also have to try Jev that, BTW, got access pretty quickly, less than a day I think...

10m agoHN ↗

OP's point here is that the overall approach of restricting output token space and using parallel prompts to produce concurrent results and taking the most relevant ones isn't something novel to Jev (not saying there's nothing novel, but a facsimile can be created at the application layer using any small, fast model)

23m agoHN ↗

Correct me if I'm wrong but Jev itself works pretty much the same as encoder only models.

17m agoHN ↗

I gave it a choice of "Foo" and "Bar" and it scored "Foo" at 98% percent. Why not 0% for both?

14m agoHN ↗

Because it's forced to rate them, there's should be a separate uncertainty parameter for both.

15m agoHN ↗

I really hate the way that LLMS design websites.

12m agoHN ↗

Unfortunately huggingface.co is blocked by my company's firewall and VPN so it breaks when downloading a model.

Are there any huggingface mirrors out there?