Home | Notifications | New Note | Local | Federated | Search | Logout

Note Detail

lain@lain@fediffusion.art (2026-09-24 12:28:28)
experimenting with longer generations... could work well with more oversight.
---Attachments---
video: https://fediffusion.art/media/71/92/65/71926581cee1d604ced087d37d4046dc828af527baddb3b6590b55fab8c9a58d.mp4

---Replies---

Kaia@kaia@pleroma.soykaf.com (2026-09-24 12:30:11)
@lain this looks very cool!

I tried to setup your video generation pipeline, but AI advised me it took 100GB for the model and space for the output and I only have 100GB free :sck:

underproofed ganbold@ganbold (2026-09-24 12:31:56)
2 mins!?

how is the inference time at your end, lain? trying to saving up to bring home H3