Viewing profile — g58892881

Thank you

Photo upscaling and inpainting apps, both running in the browser https://crisp.photos https://wipe.photos

pi

also, i double checked, nano is nano and micro is micro. didnt fuck that up

unable to fix it. latest dev ort, dispoing tensors, restarting the session/worker. nothing, crashes every time. defaulting to wasm on ios devices now

right. happens on my 13 too. memory leak confirmed, not sure yet what's causing it

wrote a web-demo, runs in your browser https://inflect-tts.geronimo-labs.com code: https://github.com/geronimi73/inflect-tts

afaik controls random noise injected into the model's latent vector before decoding same seed + same variation + same text -> bit-identical audio

well done! unet weights are in fp32. did you by any chance try something lower, fp16?

An all time classic, a headshot generator: https://instant.photos

https://crisp.photos In-browser 4x image upscaling. Vite + Onnxruntime + https://huggingface.co/Kim2091/UltraSharpV2

Surprise

pretty basic ``` from torchvision import models # Avoid downloading pretrained weights; we load trained checkpoint weights. model = models.resnet18(weights=None) model.fc = nn.Line…

I created a demo running in the browser, on your device: https://next-voice.vercel.app

my guess it will always be "some" some photos made by AI some books written by AI some code written by AI not all