Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

research.meta.ai

Post reply on HN
Print view View original

171–180 of 682 posts

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#171

Post by ericmay »

avaer wrote: post #101

I lament the comments saying this in any way redeems Meta (the company). The researchers releasing this stuff have almost nothing to do with Meta other than being bankrolled by the slaughterhouse. You aren't the customer, you are the pawn in big tech's game of thrones. Your good will is a commodity to be traded, almost literally. It will be used against you the moment it's convenient. This is open weights because Met…

It’s rather amusing to me to read comments like this, and then simultaneously whenever a Chinese company or team releases open-weight models or whatever there is a giant round of applause, America is so behind, and there’s nothing but positive things to say about the intelligent, creative, and well-intentioned Chinese engineers (which is true, America certainly doesn’t have a monopoly on great people). Don’t you know? Only China can release good, open weight models and American companies can’t compete. Oh by the way all the spend is for nothing because China alone can release open-weight models thus destroying American AI.

When an American company does anything? Doom. And. Gloom. The engineers? Taken to the slaughterhouse! America? Behind! The public? Bamboozeled!

> This is open weights because Meta couldn't monetize it in any other way than to cloud developer's judgement of their reputation.

I’ve been told over and over this doesn’t matter. Just needs to be cheap and open. Or maybe that’s only when Chyna is involved?

Sorry this post is a bit snarky but it really is something to behold. And certainly I don’t know the OP’s opinions on Chinese open weight models. Perhaps they agree with me.

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#172

Post by dev_daftly »

bentt wrote: post #65

Meta seems like the one American bigtech that would distill the the other American frontier models. My enemy’s enemy is my friend?

You think the company buying up all the books, cutting off the bindings, and feeding them through a scanner isn't also distilling other models?

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#173

Post by jszymborski »

polymorph1sm wrote: post #84

Some interesting findings from the chat template designs: 1. The template name is Onyx ATEM as found in the tool call exception message 2. It appears to be following a harmony-style chat template. But the tool use seems to be a xml like : / / 3. atem: a internal joke of meta in reverse? https://huggingface.co/meta-models/Muse-Glimmer-30B/blob/mai...

Likely inverted "meta" to avoid collision with HTMLs meta tags

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#174

Post by Der_Einzige »

cmiles8 wrote: post #97

With the business model for API based LLMs looking iffy at best it seems like we’re heading back to the “server under your desk” era of IT again.

With how expensive consumer hardware is and will continue getting (due to LLM demand), good luck getting a "server under your desk" for something less than an arm, leg, and first born.

Until A100 prices are reliably under 1.70$ an hour, there is no GPU/AI bubble and Michael Burry doesn't know anything about GPUs.

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#175

Post by drob518 »

avaer wrote: post #101

I lament the comments saying this in any way redeems Meta (the company). The researchers releasing this stuff have almost nothing to do with Meta other than being bankrolled by the slaughterhouse. You aren't the customer, you are the pawn in big tech's game of thrones. Your good will is a commodity to be traded, almost literally. It will be used against you the moment it's convenient. This is open weights because Met…

If it’s open, do you care so much that it’s from Meta? At least it should be able to give you an honest answer about Tiananmen Square.

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#176

Post by Der_Einzige »

jurgenburgen wrote: post #87
neuroticnews25 wrote: post #71

Earlier quoted context omitted.

Don't forget about energy usage, you'll probably never break even vs same model on openrouter.

If you can’t do it cheaper on your own hardware it does make you wonder how much of the cost of inference those large LLM providers are eating? Datacenter hardware isn’t magic.

Datacenter hardware might as well be magic compared to consumer. "Oh the F35 isn't magic compared to my M16 bro!"

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#178

Post by drob518 »

mycall wrote: post #64
pu_pe wrote: post #46

Earlier quoted context omitted.

Based on the benchmarks, it seems that Muse Glimmer barely edges out against Qwen3.6 27B, except for tool-calling skills (MCP, etc.). I wouldn't be surprised if they released it now because they are afraid they wouldn't beat Qwen3.8 27B.

Do AI companies make release plans based on upcoming other models like this? I would think all the processes that go into the repository and weight infrastructure pre-training, checkpointing, knowledge distillation, model compression, post training pipeline, ecosystem integrations, inference API, benchmarking, human eval/safety/alignment, docs, etc... all that dictates the release schedule.

Any company working in a competitive industry is generally aware of what their competitors are doing. PR is an important aspect to market success, so it factors into release schedule. It may not be the dominant factor given engineering constraints, but yea, it’s certainly a factor, and a large one at that.

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#179

Post by venusenvy47 »

dannyw wrote: post #150
karimf wrote: post #44

Earlier quoted context omitted.

Yes, and also waiting for the next iteration of Gemma. Muse or Qwen are optimized for coding, while IMO Gemma is still better for non-coding tasks. https://x.com/osanseviero/status/2086107547535122767

You can partially tell by the tokeniser; which gives you some hint into the training corpus mix. is four Gemma4 tokens, but one Qwen3.6 token.

Where do you find this information for each model?

Re: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows

#180

Post by drob518 »

cmiles8 wrote: post #97

With the business model for API based LLMs looking iffy at best it seems like we’re heading back to the “server under your desk” era of IT again.

That’s part of it. There’s also just a natural back-and-forth between what I call “time sharing” and “personal.” When the thing you want is expensive, you share it remotely, but as soon as costs fall, everyone wants it under their desk.
Post reply on HN