@cadey

nix shell nixpkgs#conda

should be

nix shell nixpkgs#conda nixpkgs#git

conda is looking for git for me :flan_shrug:

@cadey it's working for me! Thanks :flan_heart:

generated with sampling = 80

@solene @cadey Now I’d love to see a photorealistic doppleganger monk from Slashem running the liquid leap tecnique against a xorn.

@cadey

Thank. It's exactly what I was looking for. And far easier than what I expected. I'll try that as soon as I can, hoping it work on my laptop's GTX 1650 (I mean, it was able to train some simple custom GAN and speech syntesis).

Then I'll try to package it. (but seeing how I have an hard time getting focused on a single thing...)

@cadey
Conclusion: 4GiB doesn't work, but CPU inference seems to work. I think I mixed up the weight, though.

@cadey
(4GiB with full precision, as there are issue with 16bit floats)

Sign in to participate in the conversation
Manechat on Mastodon

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!