Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Ask HN: Is it impossible to disable Siri on macOS 27?
    discuss
  2. Show HN: JevGeni, the fruit-fly claw machine overator(jevgeni.vercel.app ↗)
    discuss
  3. Planly(github.com/drkokorev ↗)
    1comments
  4. There is no way to get ChatGPT to stop injecting your location in context(chatgpt.com ↗)
    discuss
  5. OpenAI Codex Is Retiring the Last Model That Behaves Like an Assistant(vincentschmalbach.com ↗)
    discuss
  6. The Campaign for North Africa(wikipedia.org ↗)
    discuss
  7. Disney Hires Character.ai Chief to Drive Its Technology Push(nytimes.com ↗)
    1comments
  8. The Destruction of the Scottish Canon(asteriskmag.com ↗)
    discuss
  9. AI is becoming too interconnected to fail(axios.com ↗)
    discuss
  10. Show HN: Callwitness – record what your AI agent's tools return(github.com/aditichaudharyy14 ↗)
    discuss
  11. Show HN: M0saic – deterministic video compiler (CLI, web, desktop)(m0saic.io ↗)
    discuss
  12. Is JEV just clever rule-based programming?
    discuss
  13. ChatGPT Plus and Pro Will Get Ads in 2027(vincentschmalbach.com ↗)
    1comments
  14. What I believe about the future of software development(thorstenball.com ↗)
    discuss
  15. Recreate Eliza Using JEV?
    discuss
  16. Edit command line tool – DOS Edit is back(learn.microsoft.com ↗)
    1comments
  17. RFC-0285: The Magma GPU open-source project(fuchsia.dev ↗)
    discuss
  18. Tannakian Reconstruction in a Yoneda Equipment(bartoszmilewski.com ↗)
    discuss
  19. Creating a Blog in Gemini(brennan.day ↗)
    discuss
  20. Do you use:visited on links?(kevquirk.com ↗)
    discuss
  21. NATS second glitch this month causes more disruption(bbc.co.uk ↗)
    2comments
  22. Show HN: Lossless-memory – a personal AI memory that never summarizes(github.com/aru-labs ↗)
    discuss
  23. Show HN: SQLBraid – Write SQL directly in TypeScript without a query-builder DSL(github.com/clickin ↗)
    discuss
  24. The Roadmap to Mastering LLM Inference Optimization(machinelearningmastery.com ↗)
    discuss
  25. Will new grid rules put the future of UK datacentres at risk?(computerweekly.com ↗)
    discuss
  26. He was probably right: we should be confident in our beliefs, but never certain(aeon.co ↗)
    1comments
  27. GeoJev, ask questions about the world and get its heatmaps(geojev.vercel.app ↗)
    discuss
  28. Do people prefer traditional architecture?(worksinprogress.co ↗)
    discuss
  29. Want to have a 24/7 colleague for your recruitment?(sourcegeek.com ↗)
    discuss
  30. Anthropic Principle(wikipedia.org ↗)
    1comments

Brio mode in Colibri: Scoring a closed set instead of generating

1 pointsby 1h agogithub.com
1 comments
1h agoHN ↗

Hi everyone, I’m Vincenzo, the founder of Colibrì. Today, we are adding a new mode to Colibrì called "Brio," accessible via both the TUI and the web interface. We were all impressed by Jev and its potential, so I decided to create a mode that allows you to use any model supported by Colibrì—in both chat and Brio modes. Brio returns a probability for each choice and the entropy of that distribution. No output tokens are generated, but inference work still takes place: the engine processes the context and scores the provided option tokens. You can use it via the dashboard, the terminal, or the POST /v1/brio endpoint.