search

showlab / Show-o

Repository for Show-o, One Single Transformer to Unify Multimodal Understanding and Generation.
https://arxiv.org/abs/2408.12528
Apache License 2.0
1.01k stars 44 forks source link

GPU #5

Closed yuzhongruicn closed 2 months ago

yuzhongruicn commented 2 months ago

Hi, thanks for the great work.

I got OOM on 24G 4090 GPU while trying to run the extrapolation demo. What is the minimum GPU vram required for this task?

Thank you in advance!

Sierkinhane commented 2 months ago

Hi, you can set the batch size as 1 to avoid OOM.

python3 inference_t2i.py config=configs/showo_demo.yaml \
batch_size=1 \
guidance_scale=1.75 generation_timesteps=16 \
pretrained_model_path=./checkpoints/showo.bin \
mode='extrapolation' extra_direction='left *** left *** left *** right *** right *** right' offset=0 prompt='a serene natural landscape featuring a clear, blue lake surrounded by lush green trees. *** a serene natural landscape featuring a clear, blue lake surrounded by lush green trees. *** a serene natural landscape featuring a clear, blue lake surrounded by lush green trees. *** a serene natural landscape featuring a clear, blue lake surrounded by lush green trees. *** a serene natural landscape featuring a clear, blue lake surrounded by lush green trees. *** a serene natural landscape featuring a clear, blue lake surrounded by lush green trees.' \
image_path=./inpainting_validation/alpine_lake.jpg
yuzhongruicn commented 2 months ago

Thanks