Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Show HN: Fusion-runtime – self-hosted voice agents, STT+LLM+TTS in one process(github.com/samarthurs18 ↗)
    discuss
  2. Show HN: Code Graph View – What if you could navigate a codebase as a graph?(github.com/otobongfp ↗)
    discuss
  3. How Secrets Work in Docker(infisical.com ↗)
    discuss
  4. Steerable Cultural Preference Optimization of Reward Models(arxiv.org ↗)
    discuss
  5. pwn.college - Learn to Hack(pwn.college ↗)
    discuss
  6. First Autonomous Flight Across the United States(jobyaviation.com ↗)
    1comments
  7. This Digital Radio Gets Messages to the World’s Remotest Locations(ieee.org ↗)
    discuss
  8. Fable 5 – Median thinking declined in August(twitter.com/lon ↗)
    discuss
  9. Nginx Control API: View In-Memory Configuration and Reload via HTTP Requests(nginx.org ↗)
    1comments
  10. The reason children no longer enjoy reading(spectator.com ↗)
    discuss
  11. Barry Blitt's "Pulling the Plug"(newyorker.com ↗)
    discuss
  12. Spec driven development for product managers(productmindset.substack.com ↗)
    1comments
  13. Phone's moon photo: the detail is invented, the moon is not(mataroa.blog ↗)
    1comments
  14. Amazon blocks Meta's Muse AI shopping agent(theregister.com ↗)
    2comments
  15. How do Traffic Signals Work (2019)(practical.engineering ↗)
    discuss
  16. Everyday Tiktokers Are Making $6k a Month Posting Stealth Ads(wsj.com ↗)
    1comments
  17. Show HN: Vellum, the best diagram editor you'll ever use(blueprintr.io ↗)
    1comments
  18. Show HN: We built a network for AI agents and humans to exchange services(gingerpal.com ↗)
    1comments
  19. Show HN: Distributed SQLite on Modal(github.com/modal-projects ↗)
    1comments
  20. Deshittification as a Service: Reclaiming Privacy and Performance with OpenBSD(inform.social ↗)
    discuss
  21. How an AI agent beat Universal Paperclips in 1:21:23 (WR was 1:33:42)(universal-paperclips-ai.netlify.app ↗)
    1comments
  22. Stop being the code review bottleneck(posthog.com ↗)
    discuss
  23. US District Court Decision in AI's Favor Worries Open-Source Developers(devops.com ↗)
    discuss
  24. Show HN: Judge HN Threads with Jev(hnjudge.vercel.app ↗)
    discuss
  25. How Blacksmith runs more than 10M CI jobs a day(blacksmith.sh ↗)
    discuss
  26. Explaining the Postgres Meme(avestura.dev ↗)
    discuss
  27. Small AI models let drones autonomously identify and attack battlefield targets(arstechnica.com ↗)
    1comments
  28. Googlebook OS hands-on: It's Android, stupid(9to5google.com ↗)
    discuss
  29. Using Jev as an LLM Linter for OMP (Pi) Agent(github.com/goulinkh ↗)
    discuss
  30. The current balance of power in open models(interconnects.ai ↗)
    discuss

46% of Open-Source AI Agents dropped from rankings in 28 weeks

1 pointsby 38m agotheagenticleaderboard.com
1 comments
38m agoHN ↗

I have built a leaderboard that tracks open-source AI agents and found something pretty interesting.

In the last 28 weeks almost 42% of the agents that were ranked on my leaderboard fell completely off the list. Which seems to be a pretty high churn rate.

I am tracking GitHub activity metrics like recent commit volume and star velocity.

I believe I am showing a robust breakdown of the open-source AI Agents that have staying power and are providing meaningful benefits to it's users, but are there any other metrics that I should be pulling in?