Hacker News Reader

Comment by tehsauce | Hacker News Reader

Very cool! I just released a tutorial on how to do this using the computerender api!

https://github.com/computerender/tutorials/tree/main/python/...

If you’re interested in saving money on expensive cloud gpus, our api is much cheaper than this. (only $0.001-0.0025 per frame)

View on HN

How is it so cheap? I think that's a factor of 3 cheaper than others.

The GPUs are rented from vast.ai The individual machines aren't as reliable or well-integrated with other cloud services as typical cloud machines, but multiple can be put behind a queue to create a highly reliable service.

looks cool. somehow the website looks screwed up in my browser (chrome) and I cannot get an API key after signing up.

Hm, what device/os are you browsing on? The site should be mobile-friendly except for the account page. Also please feel free to reach out by email or discord.

do you plan on adding dreambooth? I would give it a try.

I would like to support dreambooth, if there is a way to store the fine-tuned models more efficiently. The challenge is that each trained model is quite large and a bunch of models can't fit into one gpus memory at once.

Try the Lora fine-tuning method with few MB results:

https://github.com/cloneofsimo/lora

This looks great! Thanks for sharing.

textual inversion maybe? lightweight embeddings and I haven't seen any API offering it at the moment.

Reply on news.ycombinator.com