Scream Skellygore

It's not going to happen any time soon, use latent-sr to upscale your image: https://replicate.com/nightmareai/latent-sr

This didn't work on an RTX 4080 with 16GB of VRAM. `torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 16.00 GiB. GPU 0 has a total capacty of 15.99 GiB...

I would also like to see a 2x upscale as opposed to a 4x. We can always upscale in Stable Diffusion if we need higher than 512x512.

I tried your code snippet and a (screaming?) voice emanated from my Command Prompt, which I basically never, ever want to happen. Here's an alternative that uses pydub to save...

I am having the same issue.

I wasn't able to get this working on Windows 10. Here's the error I get: ```(env) C:\Users\Tristan\Documents\!Hugging Face\moondream>python sample.py --image input.png --prompt "Who is this?" C:\Users\Tristan\Documents\!Hugging Face\moondream\env\lib\site-packages\transformers\utils\generic.py:441: UserWarning: torch.utils._pytree._register_pytree_node is...

> Fyi > > pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu121 > > Change device_map to cuda on text_model.py file. > > **_Change cpu_** > > ``` > class...

I am getting this error as well, with fresh installs of webui_forge_cu121_torch231 and webui_forge_cu124_torch24. Environment: ``` python -c "import torch; print(torch.__version__)" 2.4.0+cu124 ``` ``` nvcc --version nvcc: NVIDIA (R) Cuda...