Hacker News new | ask | show | jobs
by behohippy 1099 days ago
Hey emad, thanks for SD and this! What's the plan if Meta does Apache 2.0 for LLaMA? Just keep going and making the 30b and 65b or build different models?
1 comments

Had a nice chat with Yann last week, we will release complementary stuffs.

I don't think 30b and 65b are useful given what we do, the key is optimising models for consumer & swarming them.

As for SD.. Maybe try the bot on the discord server testing the new version: https://discord.com/invite/stablediffusion

30b 4bit is definitely useful because it runs on 3090s/4090s. You see a lot of people running 30b models.

The jump in quality is very significant as well.

65b is definitely a lot less common.