How to use nvidia/audio-flamingo-next-hf with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForSeq2SeqLM processor = AutoProcessor.from_pretrained("nvidia/audio-flamingo-next-hf") model = AutoModelForSeq2SeqLM.from_pretrained("nvidia/audio-flamingo-next-hf", device_map="auto")
Hello,
Could we get an FP8 version of the model for less VRAM usage?
Thank you!
· Sign up or log in to comment
