Hacker News new | past | comments | ask | show | jobs | submit
solarkraft | on: Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows
Wow, Meta is back (at least for now)!

I like this class of model. Multi-token prediction makes it viable to run dense models at not-too-far-off speeds as MoE models with much better intelligence.

The submission’s title (open weights 30B local coding model) is luckily wrong: This is meant to be a general agentic model.

It even comes pre-quantized and with a MTP/drafter model. Looking good!

Let’s hope they aren’t dishonest with the benchmarks this time …

akazantsev | parent | next
> The submission’s title (open weights 30B local coding model) is luckily wrong: This is meant to be a general agentic model.

https://xcancel.com/alexandr_wang/status/2086756152034066792

It's correct. See the OpenCode demo. Generic models are good enough for coding without necessarily being designed specifically for coding.

solarkraft | root | parent
Right, so it's as correct as me claiming it to be an E-Mail sorting model. It may be good at that, but that's not its primary purpose.
bwfan123 | parent
> It even comes pre-quantized and with a MTP/drafter model

Glad to see the extra engineering effort that went into creating this local model and making it run well on a consumer device. I use qwen3.5-coder, and am waiting to kick the tires on this one. I hate to say this, but kudos to Meta ! I hope apple and others follow suit and create similar local models for other use cases like audio, images and video that can run on a laptop.