The smallest edge AI device for local LLMs | Hacker News Reader

The smallest edge AI device for local LLMs

tiiny.ai

22 pointsalex-moon18 comments

This entire page (except for the footer) is made with images. Is there any reason they would do this? Are they making these images by hand or are they generated somehow? Just seems like an unusual choice.

Maybe it's supposed to be less readable by AI but I really think it just makes it a little harder for humans to scrape it.

So weird. There is some compression going on too, which makes it all at least a little ugly.

I don't understand the value proposition here.

This device is selling for $1,999. In the product page on their website, they show the device plugged into a MacBook Pro.

A MacBook Pro already has excellent ability to run local LLMs, so why wouldn't someone just spend that $1,999 on maxing out their MacBook specs (M5 Max with extra RAM) instead?

Or why not buy a Mac Mini with 48GB of RAM for $2300?

> why wouldn't someone just spend that $1,999 on maxing out their MacBook specs

Perhaps they don't want a new/upgraded machine or prefer a more portable solution.

> Or why not buy a Mac Mini with 48GB of RAM for $2300?

Because this device has 80 GB, which allows running larger models (which is the whole point of the device).

AFAIK, the next significant step in terms of quality above 32 GB is 256 GB, but 80 GB still buys convenience.

> A MacBook Pro already has excellent ability to run local LLMs, so why wouldn't someone just spend that $1,999 on maxing out their MacBook specs (M5 Max with extra RAM) instead?

80 GB extra (48 → 128) cost as much as this device, but the base MBP is very expensive already (4000+$). This device is essentially a "cheap" addon for an existing system.

Pretty decent but needs more details. Overall I give it a thumbs down. You can buy a 64GB Macbook Pro M1 Max in pretty good condition right now for a good bit less than 2K.

In practice you are better off:

- if youre in the mac ecosystem using 2k and upgrading to the ultra class chip next time you want to buy

- if youre into windows wait for the next ryzen ai max chip slated for next year.

- if you want desktop consider gx10 given that you need a computer to use this anyway

In retrospect it is crazy how good the m1 max series is given the amount of ram and memory bandwidth.

I hope for their sake they bought all the RAM they needed to fulfill those Kickstarter orders at least six months ago.

The kickstarter comments page of this product does not look so glamorous ( https://www.kickstarter.com/projects/tiinyai/tiiny-ai-pocket... )

Compatible with Mac and Windows, but no Linux support? That's an odd choice. I would have picked Mac and Linux and de-prioritized Windows as a nice-to-have when there's time to get around to it.

Thought the 80GB was a bit odd, and noticed the picture of a SoC + a M.2 module:

- CIX P1/CD8180 + 32 GB RAM with 30 TOPS integrated NPU

- Houmo M50 + 48 GB RAM with 160 TOPS INT8

Anyone know more?

It's a good step in the right direction. Hardware reliability and testing needs to happen. Tell me how long it can run without heating up.

Ancient models, at < 30t/s, for $2,000? Neat device but I'll have to pass.

I am confused as to weather this is software or a product

both, but def hardware

> SoC (Armv9.2

> CPU + NPU) 30TOPS +

> dNPU 160TOPS

The Kickstarter page makes that a little more obvious - it's hardware: https://www.kickstarter.com/projects/tiinyai/tiiny-ai-pocket...

I think it's basically an NPU hanging off a USB connection, so it can be used as a compute device for local LLM inference instead of your existing CPU/GPU.

Is it just me, or others are also weirded out with that website being all images including all the text

Reply on news.ycombinator.com