←back to thread

Efficient high-resolution image synthesis with linear diffusion transformer

(nvlabs.github.io)
220 points Vt71fcAqt7 | 3 comments | 16 Oct 24 14:56 UTC | HN request time: 0.406s | source
Show context
cynicalpeace ◴[17 Oct 24 13:30 UTC] No.41869502[source]▶
None of this means much to me unless I can actually use it. Sorta like how Sora has been totally overshadowed by Kling, Runway, Minimax.

You have to release your model in some fashion for it to be impressive.

replies(2): >>41869889 #>>41870298 #
1. Agentlien ◴[17 Oct 24 14:59 UTC] No.41870298[source]▶
On the subject of such high quality video synthesis: have there been any such models which are actually available online? It strikes me that for image synthesis there have been a lot of amazing local models, but I can't remember seeing anything impressive for video which can be run offline.
replies(2): >>41870358 #>>41871436 #
2. cynicalpeace ◴[17 Oct 24 15:04 UTC] No.41870358[source]▶
The highest quality ones I mentioned are available via API or web client, but that's enough for me to be happy.
3. bick_nyers ◴[17 Oct 24 17:00 UTC] No.41871436[source]▶
CogVideoX seems to be the best offline model so far