diffusers/pr_13966/en/api/models/transformer_joyimage.md

JoyImageEditTransformer3DModel

The model can be loaded with the following code snippet.

from diffusers import JoyImageEditTransformer3DModel

transformer = JoyImageEditTransformer3DModel.from_pretrained("jdopensource/JoyAI-Image-Edit-Diffusers", subfolder="transformer", torch_dtype=torch.bfloat16)

JoyImageEditTransformer3DModel[[diffusers.JoyImageEditTransformer3DModel]]

JoyImage Transformer model for image generation / editing.

Dual-stream DiT architecture with WAN-style conditioning embeddings and custom rotary position embeddings.

• hidden_states (torch.Tensor of shape (batch_size, num_channels, num_frames, height, width) or (batch_size, num_items, num_channels, num_frames, height, width)) -- Input hidden_states.
• timestep (torch.LongTensor) -- Used to indicate denoising step.
• encoder_hidden_states (torch.Tensor, optional) -- Conditional embeddings (embeddings computed from the input conditions such as prompts) to use.
• return_dict (bool, optional, defaults to True) -- Whether or not to return a ~models.transformer_2d.Transformer2DModelOutput instead of a plain tuple.

The JoyImageEditTransformer3DModel forward method.

Transformer2DModelOutput[[diffusers.models.modeling_outputs.Transformer2DModelOutput]]

• sample (torch.Tensor of shape (batch_size, num_channels, height, width) or (batch size, num_vector_embeds - 1, num_latent_pixels) if Transformer2DModel is discrete) -- The hidden states output conditioned on the encoder_hidden_states input. If discrete, returns probability distributions for the unnoised latent pixels.

The output of Transformer2DModel.