How to use stepfun-ai/Step-Audio-R1.1 with Transformers:
# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("stepfun-ai/Step-Audio-R1.1", trust_remote_code=True, dtype="auto")
The inference example codes seem to output text only...
Yes, it output text only now, you can use a tts model to generate the audio. We will support think and output audio in step audio 2.5/3.
· Sign up or log in to comment
