alphatozeta

its really fast and the performance is great, but its really unfortunate it requires torch>=2.11 Too many NVIDIA libraries are still using 2.10 or an alpha version of 2.11 that doesn't have c++ methods used by torchcodec's underlying C++ code like use_blob and a few others. I had to fall back to ffmpeg-python unfortunately

That's awesome! congratulations!

tried omargchy, went back to vanilla hyprland

basically an advert for chroma, the vector db company, since this is essentially running a semantic similarity search. wouldn't be surprised if they're running something like clip-interrogator or just clip itself and then an approximate vector similarity search over the database of image, vector pairs in their dataset.

fine tuned custom models, models with IP knowledge, models that know what you look like. Better latency etc etc. Obviously some can be served by models hosted locally. You can host a model with Triton and create an API to call it in your native application.