MVDream-HF
A huggingface implementation of MVDream with 4 views, used for quick one-line download. See huggingface repo that hosts sd-v1.5 version and huggingface repo for sd-v2.1 version. See github repo for convertion code.
Note that the original paper presents the sd-v2.1 version. Images above are generated with sd-v2.1 version.
Convert Original Weights to Diffusers
Download original MVDream checkpoint through one of the following sources:
# for sd-v1.5 (recommended for production)
wget https://huggingface.co/MVDream/MVDream/resolve/main/sd-v1.5-4view.pt
wget https://raw.githubusercontent.com/bytedance/MVDream/main/mvdream/configs/sd-v1.yaml
# for sd-v2.1 (recommended for publication)
wget https://huggingface.co/MVDream/MVDream/resolve/main/sd-v2.1-base-4view.pt
wget https://raw.githubusercontent.com/bytedance/MVDream/main/mvdream/configs/sd-v2-base.yaml
Hugging Face diffusers weights are converted by script:
python ./scripts/convert_mvdream_to_diffusers.py --checkpoint_path ./sd-v1.5-4view.pt --dump_path . --original_config_file ./sd-v1.yaml --test
python ./scripts/convert_mvdream_to_diffusers.py --checkpoint_path ./sd-v2.1-base-4view.pt --dump_path . --original_config_file ./sd-v2-base.yaml --test
- Downloads last month
- 32
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no pipeline_tag.