How to use Composio/Mixtral-TensorRT with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Composio/Mixtral-TensorRT", dtype="auto")
Do you plan to create a FP8 PP version ?
Yes.
@SohamCom great news. Do you have a date in mind ?
· Sign up or log in to comment