celso

8 ms·

31.
by celso 3y ago
Ok, changed it to 99.8% :)
32.
by celso 3y ago
Sending you on a train to nowhere in deep Douro wasn't enough to prevent you from returning.
33.
by celso 3y ago
Best way to stay in the loop is to join their Discord server: https://discord.gg/A4z4ccJCep
34.
by celso 3y ago
You should come next year!
35.
by celso 3y ago
Less Second Life too :)
36.
by celso 3y ago
I hear this so much, sometimes in the most unexpected situations, and it's heart-warming. One day, I was having lunch with the family in a remote place in a restaurant in Serra da Estrela, and a random folk came to the table just to te
37.
by celso 3y ago
Definitely a lot of Codebits first edition vibes!
38.
by celso 3y ago
100% agree. Elect Afonso Muralha for its curator.
39.
by celso 3y ago
oh, wow, Hacker News. Inércia 2023 was raw and pure; I loved it. My post is just a tiny POV fragment of many other things that were going on there. I hope the team organizes Inércia 2024 because I'm already dusting off my 6510 assembly
40.
by celso 3y ago
Workers AI team here, glad to see demos using our APIs, happy to hear feedback.
41.
The Pictorial C64 Fault Guide(pictorial64.com)
2 points by celso 3y ago|0 comments
42.
by celso 3y ago
Or minitel...
43.
by celso 3y ago
did you see we have lamma2-7b int8?
44.
by celso 3y ago
We do support Puppeteer https://developers.cloudflare.com/browser-rendering/platform...
45.
by celso 3y ago
What do you think is missing in the docs?
46.
by celso 3y ago
The models in our catalog are all pre-loaded before requests come in.
47.
by celso 3y ago
Which models would you find useful?
48.
by celso 3y ago
We'd love to see it!
49.
by celso 3y ago
https://github.com/cloudflare/workers-wonnx
50.
by celso 3y ago
The demos in this page are using Workers AI, not WebGPU APIs. However, yes, you can perfectly run an ML inference runtime that takes advantage of GPUs using just Wasm and WebGPU APIs. That's exactly what we have here: https:/&#x2
51.
by celso 3y ago
You can use Javascript or Wasm to interface with the AI binding, think of it of as the SDK, but the inference task itself runs natively on top of a ML runtime and the models are loaded into GPUs.
52.
by celso 3y ago
On top of our hosted and supported catalog of models, and the deploy to CF partnerships like the HF one, you will also be able to bring your own custom model at some point in time.
53.
by celso 3y ago
https://blog.cloudflare.com/securing-memory-at-epyc-scale/ https://blog.cloudflare.com/speeding-up-linux-disk-encryptio...
54.
by celso 3y ago
Image generation is coming.
55.
by celso 3y ago
Unlike the first version of Constellation, Workers AI runs inference directly on GPUs that we are (quickly) installing in our global network.
56.
The West Chicago Tower Mistery(sniperinmahwah.wordpress.com)
3 points by celso 3y ago|0 comments
57.
Retro TV: Live 24/7 Vintage Cable TV Network on Raspberry Pi [video](youtube.com)
3 points by celso 3y ago|0 comments
58.
IMDB for the Terminal(github.com)
2 points by celso 3y ago|0 comments
59.
by celso 3y ago
I ordered a Plipbox Deluxe for my Amiga 1000 and it arrived today, just in time for celebrations.
60.
by celso 3y ago
This is illegal in Portugal. It's also illegal to fire someone based on the termination of a specific company role (assuming the company can do it, but this is another discussion) and then next hire someone else, contractor or not, for

More ›