> The world ships more than a billion capable phones, tablets, and laptops a year, most with a chip built for exactly this work, paid for and idle most of the day. Run the model there and the economics flip: no per-call cost, no round-trip, and nothing leaves the device.
This. I run small models (>50MB) for bio-imaging/biotech applications, it feels like every README implies that you need a discrete GPU to get started. While some do, many, especially the most useful ones, do not. Sure it matters if you're also going to do fine-tuning, but I believe your typical user just wants to detect some nuclei and get some cell-body ratios.
The laptop on your desk won't be running Meta's SAM, but it has more than enough compute to crunch 100's of your H&E slides overnight.
Cellpose[0] and Stardist[1] are the two you'll see most heavily run and talked about on the forums[2]. They're classic CNN's, but it just feels like they (and any modern models) are swept up in "Go out and buy a 4090.But these guys will give you plenty of mileage before you have to reach for the bigger ones.
I used it for testing my voice agent. It was basically what I expected. Good fast model but "generic" or "vanilla" is how i would describe its personality emulation capability as. Gemma models still outperform it in that department. As far as technicals, one thing i found annoying is cash use was not that good, it missed more then i liked, i contacted support and they were fast and responsive and said they were working on that issue, maybe they solved it with 2.5? Anyways, im prolly gonna try 2.5 again see if anything different, but cant deny the speed, thats the biggest thing this company has going for this offering as if you are in the business of classical cascaded voice agent systems, latency is number one priority and this thing is fast....
latency, instruction adherence, reliable tool cools, conversationality are all in tension.
it's great when you can get a 170ms ttft. but if you have 700 ms endpointing on the stt side and 300ms ttfb on the voice side, then you haven't really made something super snappy.
Love this take. In the bio-imaging space, Napari is a great example of this. Wonderfully solid base, but extremely extensible since its just python all the way down.
There's something wonderful about having my coworker walk up with an issue, and being able to bang out a Napari plugin that solves their exact problem before lunch.
My order of 'tool escalation' usually goes:
- Can I solve their problem from napari's inline terminal?
- Can I solve it with a one-off script?
- Can I solve it with a one-off script that creates a one-off plugin interface?
- Should I add the plugin to our company-wide repo since this problem seems to occur a lot?
I'm in biotech and if I had a dollar for every engineer I've met who dreamed of or worked-at-a-place-that-tried or even tried-in-their-freetime to make a non-invasive glucometer, I'd have at least enough to buy a coffee, which is a lot.
There is an ever-growing bodycount in the NIO-GM graveyard [0], but I too hope that one day, it'll get figured out. My old roommate and good friend was T1 and monitoring one's glucose and remembering not to eat too much/little is half your life.
That is a very bad article. Most, if not all non-invasive blood glucose projects had no peer reviewed publications.
Anyway, this problem, to measure blood glucose non-invasive, is solved now. A product should be on the market very soon. And while I am not involved anymore, I once submitted an SBIR grant application to the NIH with this technology. A specific item that this technology used would be a few hundreds bucks if produced in quantities. I also wrote this in the grant application. The examiner googled this thing, and a top-notch single item for lab measurements was 50 grand. Obviously, too expensive. So the thing I learned:
An idiot and Google are a very dangerous combination. Even at the NIH.
PS: Neither the start-up, nor their technology is mentioned in the paper. All the mentioned technologies have been tried and are bound to fail in my opinion. At least the examiner gave me a very high score on "innovative".
Ditto on the lovelyness of single-file, zero-dep, standalones. In my day job I'm stripping out hundreds of megabytes from one of our main docker containers, but running into little projects like this that make me smile. Beautiful
This is awesome! For those here not familiar with Numba, it helps bridge that performance vs ergonomics tradeoff that's always existed when you reach for python over a lower level but faster lang like C or C++.
Sure you could write Cython but then you have to have a build step and make wheels for every platform you and python version. Sure numpy has gotten faster over the years but you're still hampered by the GIL.
Numba is a little magic because you get to write stuff that feels like numpy, but get literal bytecode perf.
BUT there's a cost to this, which u learned the hard way when I imported a color map extension for matplotlib recently.
I thought i was going to be importing a couple megabytes at most. But Numba+llvmlite alone is almost 100MB!
This might be a drop in the bucket in some applications but for a color map library that has only two hot paths that need to be JITed, it's excessive.
Overall though, love this achievement, and i love what's being done for in-browser (aka local-first) scientific computing!
I run some Jupyter notebooks against a local Postgres realestate db to do statistical analysis. Could I use this to just run it on a compute host without having to manually set up and activate a venv?
I love bezier curves. In community-college, it was the first time I ever encountered a subject that made me want to go do more research on my own. One of my core memories is toiling away for multiple nights when the rest of the house was asleep on my 2015 Macbook Pro, writing janky p5.js code and pressing refresh on the browser page over and over until suddenly, I started see real, beautiful curves blossom from my control points.
I just went back to dig up some old sources[0], and I can't believe this post is almost a decade old now. This guy's explainers and animations were leagues beyond any other resource I could find through Google searc at the time.
This is fucking fantastic. I wish the whole of the internet was like this. People enthusiastically sharing their own enthusiasm about learning something cool, and other enthusiasts finding it and going down their own rabbitholes. The internet is multimedia. It should be nothing _but_ people sharing knowledge. Thanks for writing this up.
I wish they had stayed this as the very first paragraph.
Maybe I'm too stupid to grok the original declaration. Does anyone have a better summation?
reply