Text-to-Audio
Stable Audio Tools
diffusion_cond

Provide a safetensor version for model checkpoints

#6
by zhangj1an - opened

Hey authors, very impressive work! The generated audio is amazing.

May I ask is it possible to provide a safetensor version of the model weights, instead of the current .ckpt format? I'm working on integrating this model to vllm-omni (https://github.com/vllm-project/vllm-omni/issues/2003) and would appreciate it a lot.

For now, I will convert it locally first. Once the official version is released, I will re-direct it accordingly.

Thanks in advance!

Sign up or log in to comment