Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Google now ships "ads" with their Android system update?
    1 pointsby 0 minutes ago
    —discuss
  2. Ask HN: What're You Reading?
    1 pointsby 2 minutes ago
    —discuss
  3. Building a certificate authority for the whole Internet (cloudflare.com)
    1 pointsby 3 minutes ago
    —discuss
  4. The Role of Theory in Biology (asimov.press)
    1 pointsby 7 minutes ago
    —discuss
  5. How to Live Better? (dlgrv.github.io)
    1 pointsby 8 minutes ago
    —discuss
  6. A challenging, contamination-free LLM benchmark (livebench.ai)
    1 pointsby 8 minutes ago
    —discuss
  7. Sculpting DTS support into a 2020 LG CX (jhm.nz)
    2 pointsby 9 minutes ago
    —discuss
  8. Using Device Linking to Eavesdrop on WhatsApp and Signal (schneier.com)
    1 pointsby 11 minutes ago
    —discuss
  9. Show HN: Check whether a job posting is still real before you apply (unlisted.careers)
    1 pointsby 13 minutes ago
    —discuss
  10. Backblaze Drive Stats for Q2 2026 (backblaze.com)
    2 pointsby 13 minutes ago
    —discuss
  11. Rust coreutils acknowledges copying of GNU GPL code (github.com/uutils)
    2 pointsby 13 minutes ago
    1comments
  12. Show HN: GoDrop – one curl turns a file into a URL your agent can paste (godrop.sh)
    1 pointsby 14 minutes ago
    —discuss
  13. Optimising for fast builds with GHC [video] (youtube.com)
    1 pointsby 15 minutes ago
    —discuss
  14. Minecraft inside a particle accelerator is weird [video] (youtube.com)
    1 pointsby 15 minutes ago
    —discuss
  15. What makes software development engineering (parksb.github.io)
    2 pointsby 16 minutes ago
    —discuss
  16. CoW – a stacking window manager for Wayland (codeberg.page)
    2 pointsby 16 minutes ago
    —discuss
  17. What's a CEO Worth Anyway? (mleverything.substack.com)
    1 pointsby 16 minutes ago
    —discuss
  18. Show HN: Find Wazuh rules that never fire, including ones dropped at load (github.com/xuxu298)
    1 pointsby 17 minutes ago
    —discuss
  19. Planly – offline-first day planner PWA, self-hosts with one Docker Compose up (github.com/drkokorev)
    1 pointsby 18 minutes ago
    —discuss
  20. Oya, an agent does a browser task once, then it replays with no LLM (oyabrowser.com)
    1 pointsby 19 minutes ago
    —discuss
  21. Meta Enterprise Platform to be led by outgoing MongoDB boss (siliconrepublic.com)
    1 pointsby 23 minutes ago
    —discuss
  22. A surprising number of Americans use AI for spiritual reasons (npr.org)
    1 pointsby 23 minutes ago
    —discuss
  23. Autonomous AI in Space Exploration Raises New Stakes (ieee.org)
    1 pointsby 23 minutes ago
    —discuss
  24. Poetry for Engineers: User Interface Design (ieee.org)
    1 pointsby 25 minutes ago
    —discuss
  25. Show HN: HankyDoodle – a Draw Something-style game I built with my kids (hankydoodle.app)
    1 pointsby 25 minutes ago
    2comments
  26. 220 Per Cent More Code: Why Meta Stopped Replacing Staff With AI (smarterarticles.co.uk)
    1 pointsby 25 minutes ago
    —discuss
  27. The Listening Room Is Having a Design Moment (design-milk.com)
    1 pointsby 27 minutes ago
    —discuss
  28. Friends embark on 20-year board game battle (bbc.com)
    1 pointsby 28 minutes ago
    —discuss
  29. The days of the online job application are OVER (recruited.org.uk)
    1 pointsby 29 minutes ago
    —discuss
  30. Protobuf Everywhere: How Lupyd Uses Protocol Buffers Across Web, Mobile, DB (lupyd.com)
    2 pointsby 29 minutes ago
    —discuss

A challenging, contamination-free LLM benchmark

1 pointsby 12m agolivebench.ai
Open original View on HN
0 comments
A quiet thread, for now.Start the conversation on HN ↗