Are there any ways to make this run faster on low-end GPUs?

VectorSpaceLab / OmniGen

OmniGen: Unified Image Generation. https://arxiv.org/pdf/2409.11340

MIT License

2.89k stars 228 forks source link

Are there any ways to make this run faster on low-end GPUs? #93

Open runshouse opened 3 weeks ago

runshouse commented 3 weeks ago

I use a T4 and am generating 640x384 images at 30 steps in 5.5 minutes. I've tried most of the optimizations that are in inference.md and that chatgpt recommends with no luck. Thanks!