Instructions to use ACE-Step/acestep-transcriber with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ACE-Step/acestep-transcriber with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("ACE-Step/acestep-transcriber") model = AutoModelForMultimodalLM.from_pretrained("ACE-Step/acestep-transcriber") - Notebooks
- Google Colab
- Kaggle
Considering releasing the FP8 quantization model?
#1
by a463724055 - opened
The weight is too large.(Cry)
You can quantize it yourself
You can quantize it yourself
ok, setting CPU offload allows it to run with 16GB VRAM.
a463724055 changed discussion status to closed