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

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 an hour ago | parent [-]

Nice work!

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

▲

Aeroi 22 minutes ago | parent | next [-]

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

▲

embedding-shape an hour ago | parent | prev [-]

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.

	▲	junofan 17 minutes ago \| parent [-]
		I asked muse. It’s for codex-cli, and it routes through “OpenAI inference proxy”

▲

Tiberium an hour ago | parent | prev | next [-]

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

▲

Aeroi an hour ago | parent [-]

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 34 minutes ago \| parent [-]
		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.

▲

6thbit 38 minutes ago | parent | prev | next [-]

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_ 23 minutes ago \| parent [-]
		No reason for it to. Software companies use products from other software companeis all the time.

▲

manav 27 minutes ago | parent | prev | next [-]

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

▲

sroussey 24 minutes ago | parent [-]

I’ve had Claude opus do that too.

▲

wincy 14 minutes ago | parent | next [-]

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.

▲

tclancy 16 minutes ago | parent | prev | next [-]

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.

▲

ChickeNES 8 minutes ago | parent | prev [-]

A Hindi word in the middle of a normal reply for me (but once I translated the word it was right in context heh)

	▲	arshxyz 2 minutes ago \| parent [-]
		This used to happen a lot with GPT 5.4. It would start outputting entire sentences in Korean for no apparent reason.

▲

GenerWork an hour ago | parent | prev [-]

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.

	▲	Aeroi 42 minutes ago \| parent [-]
		i agree, and it's what made me spend time exploring today. fair q. call_ ids and gAAAAA blobs arent damning but the rs_ reasoning ids embed a unix timestamp that matches the session to the second, then OpenAI's 819x marker. plus the summary is in OpenAI's summarizer voice. its just a best guess.