Windows AI Studio Preview

[blackoil]

Windows AI Studio is in real a Linux AI Studio as it needs WSL to run. Little funny

[severino]

But can you get this to run in Linux? From the article I can see it's a VSCode extension or something like that but the name kind of implies it's Windows only.

[mmis1000]

that probably explain why nvidia only. Amd don't really have rocm for linux on latest consumer cards.

[troupo]

Obligatory "is this the year of Linux in desktop" joke

[Jayakumark]

Adding link to VScode extension:

https://marketplace.visualstudio.com/items?itemName=ms-windo...

[davikr]

Anyone knows when Microsoft will release a local OCR model officially? I haven't seen anyone talking about this, but the one they're shipping with Snipping Tool (OneOcr) is top-tier and beats everything out there like Tesseract, easyOCR etc.

The model is technically in everyone's Windows installs, but we don't have the C++ projected WinRT headers to use the Microsoft.Windows.Vision library.

[capableweb]

Seems they've released a bunch already?

Image-to-text models, filtered by "Microsoft": https://huggingface.co/models?pipeline_tag=image-to-text&sor...

[andruby]

Apple has good uniform hardware to enable this, but they are a product company and an “AI studio” would not fit their usual definition of a product.

I do hope they are considering going in that direction though.

[paradite]

Apple has it: https://github.com/ml-explore/mlx

[TheRoque]

That's the weird thing. They seem to have the best hardware for the price, for individuals to develop and use local LLMs, but so far they have been pretty quiet on all this.

[m3kw9]

They do have such tools for developers like CreateML to train your own models, pretty sure they will have one for LLM.

[lostmsu]

Nobody is going to train or even fine-tune large models on Apple hardware. It is too slow for that purpose.

[Const-me]

> will run only on NVIDIA GPUs for the preview

I wonder why Microsoft helps nVidia, instead of using their own technology?

Here’s an example: https://github.com/Const-me/Cgml

[pjmlp]

Their own technology is DirectML, not raw DirectX 11.

https://blogs.windows.com/windowsdeveloper/2023/11/15/elevat...

https://blogs.windows.com/windowsdeveloper/2023/12/14/direct...

[capableweb]

Companies of the size of Microsoft usually function more like a country with various companies within, rather than one company with a unified view and unified goals.

So probably, the group wasn't even aware of that technology because it's far away by either professional connection, or by personal/relationship connections, or they knew about it but had another goal than "maximize use of own stuff" and made the call that the tradeoffs wasn't worth it.

[lostmsu]

Have you considered ILGPU? https://ilgpu.net/

[DeathArrow]

Microsoft also has a nice AI/ML framework called ML.NET.

[a1o]

Curious it uses Ubuntu 18.04 with 20.04 and 22.04 already released and 24.04 just around the corner.

[ourmandave]

RAG Project

Coming soon!

=(

[franzb]

Why is this on GitHub?

[Mashimo]

Why would it not be?

[mngdtt]

Installed this, and it didn't let me do anything. Despite the fact the readme says stuff runs locally, first it asked me to link to my github account, and then all models required me to ask someone (I think meta?) for permission on github or use a huggingface token or whatever.

So I uninstalled it and now my wsl prompt starts with (base) and I don't know how to disable it and all my python scripts are broken because they can't find all the libraries I've installed from pip throughout the years.

0/10 would not recommend.

[ZunarJ5]

You need to turn off the default conda environment.

Try: conda config --set auto_activate_base false

[dazzlefruit]

(base) sounds like Anaconda. Try typing conda deactivate?