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

101–110 of 682 posts

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

#101

Post by avaer »

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 Meta couldn't monetize it in any other way than to cloud developer's judgement of their reputation.

But I guess most people just don't care.

I'm glad it's open. It does not make me think any better of Meta.

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

#102

Post by darig »

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…

[dead]

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

#103

Post by flaburgan »

cynicalsecurity wrote: post #69
sajithdilshan wrote: post #10

Still needs 32-64GB memory to run it locally. 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany. A more practical model would be a language specific (e.g Python or JVM language) and excellent at tool calling and reasoning. Maybe that way they can shrink it even more.

I don't understand the desire to run own AI models for programming locally. No laptop is ever going to be as powerful and energy efficient to run anything close to OpenAI, Anthropic or Google models. A model you can run on a loptop is simply not going to work as well as it's needed for programming. Small models for linguistic work fine, but anything more sophisticated simply won't provide enough resources or power. O…

Yet.

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

#104

Post by lluisantoni »

cynicalsecurity wrote: post #69
sajithdilshan wrote: post #10

Still needs 32-64GB memory to run it locally. 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany. A more practical model would be a language specific (e.g Python or JVM language) and excellent at tool calling and reasoning. Maybe that way they can shrink it even more.

I don't understand the desire to run own AI models for programming locally. No laptop is ever going to be as powerful and energy efficient to run anything close to OpenAI, Anthropic or Google models. A model you can run on a loptop is simply not going to work as well as it's needed for programming. Small models for linguistic work fine, but anything more sophisticated simply won't provide enough resources or power. O…

For some companies there might be a need to run them locally. For instance, Apple decided to run LLMs on the phone locally. I guess it depends on how important latency and privacy are. Perhaps Meta is looking at how much interest for those local models is there.

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

#105

Post by captainbland »

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…

To be honest the main issue with meta has never been around open/closed software. They've also done react, Cassandra and some other bits. But this, like their open weights is like a feather pressing down on the scale compared to things like promoting genocide in Myanmar, enabling Cambridge analytica, creating a huge closed ecosystem which dominate(s/d) local community communication, mandating doxxed communication, trying to replace actual community communication with algorithmic nonsense etc.

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

#106

Post by eigenspace »

sajithdilshan wrote: post #10

Still needs 32-64GB memory to run it locally. 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany. A more practical model would be a language specific (e.g Python or JVM language) and excellent at tool calling and reasoning. Maybe that way they can shrink it even more.

I think if there's going to be advantages to making smaller, more targeted models, those advantages will probably come from targeting specific domains, not from targeting specific languages.

I think that if an LLM can't abstract over the differences between Python and C++, it probably will have an even harder time abstracting over the differences between writing code that manages a webserver, and writing code that does aerodynamic simulations.

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

#107

Post by Archit3ch »

sajithdilshan wrote: post #10

Still needs 32-64GB memory to run it locally. 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany. A more practical model would be a language specific (e.g Python or JVM language) and excellent at tool calling and reasoning. Maybe that way they can shrink it even more.

> 64GB Macbook pro with an M5 chip costs more than 4k Euros in Germany

Sure, if you want the latest and almost* greatest. You can pick up an M1 Max 64GB for ~1k.

* I guess 128GB also exists

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

#108

Post by petu »

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 can batch at large scale, probably over 90% more energy efficient per token than a MacBook.

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

#109

Post by larodi »

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…

Meta and its products, as a whole, is a threat to your kids, your mental health, your community's health and the planet as a whole. It is just sad and very repulsive everyone fell so easily addicted to their social drug. Yes - it is a drug, and it is hard to get off from.

Nothing redeems them at this point of time, they are doing exactly ZERO to redeem. Tossing open weight models (not opensource!!) is not a basis for redemption, and does not constitute remorse in any way. Trying to portray it as such is complicity to META's crimes against humanity.

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

#110

Post by commoner »

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…

Muse Glimmer doesn't redeem Meta, but it's a contribution to the commons and the Apache 2.0 licensing is an improvement from the restricted licenses attached to Llama. If even Meta can use a permissive license for its model weights, so can any other company.
Post reply on HN