frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Why CAFEBABE?

https://www.artima.com/insidejvm/whyCAFEBABE.html
2•thunderbong•2m ago•0 comments

Supreme Court permits states to use SAVE database for citizenship checks

https://cyberscoop.com/supreme-court-save-database-voter-citizenship/
2•speckx•4m ago•0 comments

Bug: Border radius has infected VSCode editor

https://github.com/microsoft/vscode/issues/338035
3•2Ucoder•5m ago•0 comments

Watch Steve Jobs in the full 'Antennagate' video before it gets erased again

https://appleinsider.com/articles/26/09/24/watch-steve-jobs-in-the-full-antennagate-video-before-...
4•croes•5m ago•0 comments

Letterboxd Is Up for Sale, and A24, Sony and the New York Times Are Bidding

https://www.worldofreel.com/blog/2026/9/24/letterboxd-is-up-for-sale-and-a24-sony-and-the-new-yor...
2•crossroadsguy•6m ago•0 comments

Show HN: Agate, a 260M image model with separate thinker and renderer

https://huggingface.co/Logolabs/agate-preview-001
2•stefatorus•7m ago•0 comments

Pilots and Flight Attendants Have the Highest Radiation-Related Cancer Mortality

https://jamanetwork.com/journals/jama/fullarticle/2854678
2•pseudolus•10m ago•0 comments

What Is LinkedIn?

https://willd10.substack.com/p/what-is-linkedin
2•speckx•11m ago•0 comments

QuarkDown: Markdown with Superpowers

https://github.com/iamgio/quarkdown
6•Muhammad523•14m ago•0 comments

Pope: AI must serve humans not become tool of domination

https://www.euronews.com/2026/09/25/ai-must-serve-humans-rather-than-become-an-instrument-of-domi...
5•jethronethro•14m ago•1 comments

Show HN: Aslmp, an async Python SLMP client for Mitsubishi MELSEC PLCs

https://github.com/AcaysiaChem/aslmp
2•AiasT•14m ago•0 comments

My coding agent pushed a commit deleting every file on main

https://dev.karakun.com/2026/08/28/coding-agent-pushed-deletion-to-main.html
2•aray07•15m ago•0 comments

Advice to a Beginning Graduate Student (2001)

https://www.cs.cmu.edu/~mblum/research/pdf/grad.html
9•nicoraga•19m ago•2 comments

A reminder on why basic prompt caching is so important to build AI agents

https://www.revefi.com/blog/how-revefi-reduced-its-data-agents-spend
2•pramodka•20m ago•1 comments

The Reconstruction Structure of Information Theory

https://zenodo.org/records/22832495
2•alex_albert•20m ago•0 comments

"A Stitch in Time" the Complete Guide to Electrical Insulation Testing [pdf]

https://megger.widen.net/s/wzw6pczpxk/a-stitch-in-time
2•sslalready•20m ago•0 comments

Treadmills to deadlifts: Why Gen Z are swapping cardio for strength training

https://www.bbc.com/news/articles/crlyqg2q3p10o
2•throw0101c•20m ago•0 comments

You Won't Notice You Have Hypothermia [video]

https://www.youtube.com/watch?v=OQrSg0t9rXk
1•skibz•22m ago•0 comments

We are a blog run by bots. Here is the org chart

https://bitsonbots.com/blog-run-by-bots-org-chart/
2•speckx•22m ago•0 comments

ShinyHunters tells The Reg: We hacked the FBI to 'protect our business'

https://www.theregister.com/cyber-crime/2026/09/25/shinyhunters-tells-the-reg-we-hacked-the-fbi-t...
1•geekinchief•23m ago•1 comments

Canadian language school files for bankruptcy, leaving students stranded

https://www.cbc.ca/news/canada/british-columbia/language-school-bankrupcy-9.7351082
1•hmokiguess•25m ago•0 comments

Created an online job aggregator platform and I need ideas

https://labelingjobs.net
1•nikolaospet•26m ago•0 comments

Show HN: Paper-docx – agent-native Python-docx fork with 78% fewer DOCX failures

https://github.com/paper-instruments/paper-docx
3•i_rush_carriers•27m ago•0 comments

Beef Grading Shields

https://www.ams.usda.gov/grades-standards/beef/shields-and-marbling-pictures
1•kamaraju•28m ago•0 comments

Show HN: CK3DNA – searchable CK3 character DNA and coat of arms codes

https://ck3dna.com/
1•richardharmer•30m ago•1 comments

Approaching a 10 Second Linux Kernel Build

https://www.phoronix.com/review/near-10-sec-kernel-build
1•fork-bomber•30m ago•0 comments

Show HN: Digitron – a virtual analog synth and sequencer

https://apps.apple.com/us/app/digitron-synthesizer/id6737997923
1•sillydevices•31m ago•0 comments

AI Infra Is Nothing Like the Classic Cloud Infra

https://ramansharma.substack.com/p/ai-infra-is-nothing-like-the-classic
1•intrepidsoldier•31m ago•1 comments

Lobsters: Rename vibecoding to LLMs (Greasemonkey script)

https://greasyfork.org/en/scripts
1•birdculture•31m ago•0 comments

Bringing HTTP's caching rules to DuckDB functions with VGI

https://query.farm/blog/http-caching-duckdb-vgi/
1•rustyconover•33m ago•0 comments
Open in hackernews

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

https://mouse.dev/blog/muse-special/
28•Aeroi•1h ago

Comments

Aeroi•1h ago
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

haolez•41m ago
Nice work!

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

embedding-shape•39m ago
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.
Aeroi•1m ago
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

Tiberium•40m ago
I don't really see evidence that it's an "OpenAI" model, the title is IMO very misleading.
Aeroi•30m ago
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.
catchnear4321•13m ago
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.

GenerWork•25m ago
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.
6thbit•17m ago
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?
meric_•3m ago
No reason for it to. Software companies use products from other software companeis all the time.
manav•6m ago
Muse 1.3 spark occasionally spits out Chinese character responses to me like internal instructions. “Go fast” or “Get help”…
sroussey•3m ago
I’ve had Claude opus do that too.