Hugging Face has a very large list of supported transformers. They provide a table which gives the status on whether or not a transformer has a slow tokenizer, a fast tokenizer, PyTorch support, TensorFlow support, and Flax support. Given that, and many other great informative pages, there is no easy way to determine which model, and its variants, are available given certain GPU memory requirements. For example, if I have an Nvidia Quadro RTX 8000 with 48GB of memory, which transformers can I run in inference mode?



You must log in to answer this question.

Browse other questions tagged .