How to use lapp0/esm2_speedrun with Transformers:
# Load model directly from transformers import ESM model = ESM.from_pretrained("lapp0/esm2_speedrun", device_map="auto")
