Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Ollaya – Ollama for open-source, Jev-style decision models (ollaya.dev)
    30comments
  2. Alan Kay: Shannon gave us a way of dealing with noisy channels [video] (youtube.com)
    8comments
  3. Advice to a Beginning Graduate Student (2001) (cmu.edu)
    2comments
  4. Platform-independent SIMD in Go (go.dev)
    111comments
  5. First Principles Thinking (sunilsadasivan.com)
    61comments
  6. Git-bug: Distributed, offline-first bug tracker embedded in Git (github.com/git-bug)
    83comments
  7. U.S. appeals court upholds designation of Anthropic as supply chain risk (cnbc.com)
    385comments
  8. Google's first Suncatcher orbital data center test launches October 1 (arstechnica.com)
    1comments
  9. Supreme Court permits states to use SAVE database for citizenship checks (cyberscoop.com)
    —discuss
  10. Meta's Muse appears to use an OpenAI model labeled muse-special (mouse.dev)
    17comments
  11. Pentium II at 600Mhz with Voodoo 3 Emulated on 86Box with M6 Mac Mini (nyaa.sh)
    105comments
  12. Ink and Switch interactive homepage (inkandswitch.com)
    24comments
  13. Show HN: Doom or Bloom, map your AI worldview (doom-or-bloom.com)
    20comments
  14. Factorio that you can touch (factorio.com)
    59comments
  15. Amiga Screens: A Primer (datagubbe.se)
    25comments
  16. Zelensky says Russia has widened attacks to hit Ukraine's data centres (bbc.com)
    40comments
  17. Show HN: Whiteboard (YC W26) – An open-source IDE for thoughtful software design (github.com/devdotfast)
    127comments
  18. What About Rails? (jardo.dev)
    166comments
  19. Boards of Casio (ambionix.com)
    26comments
  20. Typst makes big strides (lwn.net)
    3comments
  21. CVE-2025-13032: Entering and Breaking the Avast Antivirus Sandbox Part 2 (safateam.com)
    27comments
  22. Why is the liver so weirdly regenerative? (dynomight.substack.com)
    266comments
  23. Show HN: Hamilton – a no-internet Android health dashboard, named after my dog (play.google.com)
    10comments
  24. What happens when you analyze your favorite college football team like the CIA? (cultivatelabs.com)
    7comments
  25. Astronomer watches Starlink satellites sinking to build a 'planetary barometer' (theregister.com)
    1comments
  26. Rails World 2026 Opening Keynote [video] (youtube.com)
    459comments
  27. 2DWillNeverDie (2dwillneverdie.com)
    81comments
  28. Gravity Seems Holographic. What Does That Mean for Reality? (quantamagazine.org)
    70comments
  29. Opus 5.5 is good at explainer videos (launchvideo.io)
    210comments
  30. Toyota is taking the Corolla electric (electrek.co)
    763comments

Meta's Muse appears to use an OpenAI model labeled muse-special

31 pointsby 1h agomouse.dev
15 comments
1h agoHN ↗

author here. I kept digging through the Muse filesystem after my first post hit the front page here this week.

Going through my logs I found a background agent that used a model called azure/muse-special while building my website. I found this interesting and dug a little deeper.

The transcript and daemon binary point to an OpenAI model running on Azure. Still unclear which one or why it was selected.

The runtime also ships with an Anthropic client and a catalogue listing Claude, GPT and Kimi models. I didn’t observe Claude being used in my own sessions, but this is my Part 2 of exploring the Muse file system.

original post: https://x.com/heypeterjames/status/2103545183400800746

pete at mouse dot dev

56m agoHN ↗

Nice work!

Are you sure this isn't just a custom model that is API-compatible with OpenAI?

54m agoHN ↗

If it's using the same weird (and awful) "backend prompt encryption" pattern and mechanism that Codex also uses (https://news.ycombinator.com/item?id=48905028), then I'd also say it points to OpenAI being involved. Hopefully that shit isn't becoming a more popular pattern, absolutely awful for troubleshooting stuff.

11m agoHN ↗

I asked muse. It’s for codex-cli, and it routes through “OpenAI inference proxy”

16m agoHN ↗

I'm not no. Someone from meta would have to chime in... but its very openai shaped and served differently from all the other models listed in the daemon, under a mysterious name.

a fine-tuned OpenAI model on Azure for the purpose of compaction or something could make sense I guess but that would still be OpenAI's weights with meta ontop, and they already have a compaction model served under azure/avocado-compaction-v1

55m agoHN ↗

I don't really see evidence that it's an "OpenAI" model, the title is IMO very misleading.

45m agoHN ↗

Meta already serves its own models on Azure under their real names. azure/avocado-compaction-v1 and azure/avocado-memory-flush-v1 are in the same catalog. Those sessions do not come back as gpt_responses_v1 items with rs_ ids and encrypted reasoning.

28m agoHN ↗

router makes sense. fill a gap with someone else’s model wearing a mask, swap it out with your model not wearing a mask when (or if) you reach that.

definitely tracks with alexandr’s hand-wavey influencer turn.

40m agoHN ↗

I'm not sure I understand why they would route it to other models. It can't be that they don't have enough compute. Maybe worried that the answer from their own models would be bad? Doesn't really make sense, but I could be missing something.

32m agoHN ↗

It'd be hilarious that meta chose to pay openai. If there was such a deal, would it come to light in any public/official filings?

18m agoHN ↗

No reason for it to. Software companies use products from other software companeis all the time.

21m agoHN ↗

Muse 1.3 spark occasionally spits out Chinese character responses to me like internal instructions. “Go fast” or “Get help”…

11m agoHN ↗

Got Russian from it once. As a child of the '80s I jumped right on it, but the explanation seemed believable.

Still sleeping with one eye open.

9m agoHN ↗

I had a totally benign chat with OpenAI and it titled it as “amateur porn” in Chinese characters, it was very alarmed when I pointed the conversation name out to it. It almost never misses these days, but when it does the failure modes are very strange.