cuda memory error

[slives-lab]

The script to generate image from text works (python scripts/txt2img.py --prompt "a photograph of an astronaut riding a horse" --plms ). However, the img2img script fails with a cuda error (python scripts/img2img.py --prompt "A fantasy landscape, trending on artstation" --init-img <path-to-img.jpg> --strength 0.8).

RuntimeError: CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 15.78 GiB total capacity; 8.17 GiB already allocated; 5.91 GiB free; 8.63 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

[surister]

You don't have enough memory, try to adjust the height and width to a smaller ratio, 512x512 for example.

[slives-lab]

What is the VRAM requirement to avoid error?

[g-jing]

How to change the output image size for img2img.sh?

[andresmgsl]

How to change the output image size for img2img.sh?

I have the same question.. any help?? thanks