As long as the hardware is cheap enough and gets into the right hands people will train and research to improve open models.

Cory Doctorow wrote a blog post on this.

It depends on what you mean by improve.

Don’t mind the downvotes, they’re from people who don’t even use this community.

Calling it a “Distillation Attack” is wild. Get fucked Anthropic.

Something like this would go hard in real life.

Asagao to Kase-san, a one episode OVA 00:05:25 - 00:05:57.

I’ve updated the post. Thanks.

I couldn’t tell what tool they were using or crop they were tending, so I went with the safer option, to me. If you could help me out here it’d be much appreciated.

I don’t know, I can’t tell.

😞

I watched it a long time ago. I don’t remember what happens in it, but I remember liking it.

Yeah.

Yeah, it’s a different form of quantization. I think it’s supposed to look a little better than conventional quants, but it still has all the tradeoffs of a quant.

The license is Apache 2.0, but I don’t know how much VRAM it takes.

VAEs are used in image generation too at the end of generation to convert latent images to pixel space.

That sounds cool. How’s it do the words in the background?

I haven’t seen this one, but I always hear it’s good.

The chrome finish so simple and such an aesthetic. Already a classic by 1968.

Let’s hope this world isn’t geologically active.