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

Nativ: Run frontier open models locally on your Mac

blaizzy.github.io

Post reply on HN
Print view View original

131–140 of 144 posts

Re: Nativ: Run frontier open models locally on your Mac

#131

Post by sgt101 »

walrus01 wrote: post #80
erikkahler wrote: post #35

To clarify, this MIT-licensed app is from the very same dev, 'Prince Canuma', who maintains the popular MLX-VLM library ( https://github.com/Blaizzy/mlx-vlm ). MLX-VLM is a long-time dependency of the excellent LM Studio and others because it can provide faster inference on Apple devices than llama.cpp. Historically, MLX is a smaller community than CUDA, but has some of the fastest updates upon the release of new mod…

I would also note that for people who want to download these models, you can find MLX versions of just about everything popular on huggingface these days. For instance go look at the "main" page for Qwen 3.6 35B-A3B and then follow the link to quantizations, and pick one of the more popular/reputable MLX variants. https://huggingface.co/Qwen/Qwen3.6-35B-A3B

converting a model to MLX is trivially easy as well - just a couple of lines of code and a few minutes wait.

I do find that it reliably crashes on the first try, but then it picks up from where it finishes and completes the conversion reliably.

Re: Nativ: Run frontier open models locally on your Mac

#132

Post by IshKebab »

andsoitis wrote: post #120
IshKebab wrote: post #91

Earlier quoted context omitted.

The submission is "Nativ: Run frontier open models locally on your Mac" and on the page it says "Nativ puts frontier intelligence on your desk."

> puts frontier intelligence on your desk Which of the models it puts on your desk do you consider frontier intelligence?

None of them.

Re: Nativ: Run frontier open models locally on your Mac

#133

Post by D13Fd »

syntaxing wrote: post #78
D13Fd wrote: post #16

I'm surprised that their home page basically acts as if LM Studio and others don't already do this. It's not clear what the difference is from a glance. It also omits Open WebUI. I've been running Deepseek V4 Flash locally on my Macbook Pro for weeks using Open WebUI + DS4.

What spec is your macbook? I want to run Deepseek V4 Flash but its too slow for agents on my Strix Halo.

It’s a maxed out current-gen MacBook Pro w/ 128gb ram. I lucked out in getting it before they upped the price.

It runs really well and fast enough for what I need, and it works great, but it’s slower than Claude. I use it for projects where I’m not allowed to use third party AI for legal reasons.

Re: Nativ: Run frontier open models locally on your Mac

#134

Post by dostick »

justatdotin wrote: post #85
simonw wrote: post #77

Earlier quoted context omitted.

I was so excited when I saw "blaizzy" in the domain, because Prince Canuma's work around MLX has been of such uniquely high quality.

Hi Simon, sorry to spam your comments but 7 months ago you asked me for a media report to back a claim I made and this week it finally arrived. perhaps a little late to be very useful to anyone who maintains that gating function, but here nonetheless: https://ntindependent.com.au/scientist-says-ministers-pole-f...

What

Re: Nativ: Run frontier open models locally on your Mac

#135

Post by mips_avatar »

febed wrote: post #89
mips_avatar wrote: post #63

Earlier quoted context omitted.

Qwen35ba3b can do a huge amount of data cleaning work on pretty modest hardware. Already have run about 100 billion tokens on it using 2x3090 gpus.

What exactly do you mean by data cleaning

The coolest project I’ve got this running on is improving the depicts metadata for photos on Wikipedia. A lot of times they won’t have the landmarks tagged correctly in a photo. So I will load in all the metadata that exists from each photo and the pixels of those photos and give a small qwen agent access to Wikipedia search as well as a geocoder. It does a great job of figuring out what is depicted and tagging it with the correct depicts field. Im still early on but I have been able to double the number of places that have a photo attached to them on wikimedia

Re: Nativ: Run frontier open models locally on your Mac

#136

Post by mips_avatar »

noja wrote: post #110
mips_avatar wrote: post #63

Earlier quoted context omitted.

Qwen35ba3b can do a huge amount of data cleaning work on pretty modest hardware. Already have run about 100 billion tokens on it using 2x3090 gpus.

are you using it differently to OpenRefine?

I’ve found it helps a lot with reconciliation tasks where tools like openrefine can’t handle it. Like I wanted to tag blog posts with links to Wikipedia articles that are relevant. But the thing is whats relevant changes a lot based on context in the blog post. So like a naive reconciler will tag the article for “sky” in the blogpost title “the sky above the Notre dame shone the morning we visited” when the element that should be tagged is the cathedral.

Re: Nativ: Run frontier open models locally on your Mac

#137

Post by jboss10 »

oersted wrote: post #109
jboss10 wrote: post #105

Earlier quoted context omitted.

I think this is one of the easiest to understand. https://arena.ai/leaderboard/text/pareto

That’s a good visualization, although I am a bit mistrustful of Arena’s scores. It does get around the fact that models are getting trained for the benchmarks, but the methodology of letting random people compare outputs side-by-side is a very shallow judgement method in my opinion. EDIT: Indeed looking at the overall rankings for text again, the list is rather strange, a lot more about writing style than intelligenc…

They supposedly have a style control system, but I doubt it's perfect. I wish there was a parento view like this for agentic systems(using a standard harness)

Re: Nativ: Run frontier open models locally on your Mac

#140

Post by febed »

mips_avatar wrote: post #135
febed wrote: post #89

Earlier quoted context omitted.

What exactly do you mean by data cleaning

The coolest project I’ve got this running on is improving the depicts metadata for photos on Wikipedia. A lot of times they won’t have the landmarks tagged correctly in a photo. So I will load in all the metadata that exists from each photo and the pixels of those photos and give a small qwen agent access to Wikipedia search as well as a geocoder. It does a great job of figuring out what is depicted and tagging it wi…

Interesting project. So I guess the search tool is to crawl Wikipedia for articles? And how do you ensure that the tagging stays within the Wikidata taxonomy? How exactly are you using a geocoder? Sorry, just curious
Post reply on HN