diffusers/pr_14401/en/api/models/allegro_transformer3d.md

AllegroTransformer3DModel

A Diffusion Transformer model for 3D data from Allegro was introduced in Allegro: Open the Black Box of Commercial-Level Video Generation Model by RhymesAI.

The model can be loaded with the following code snippet.

from diffusers import AllegroTransformer3DModel

transformer = AllegroTransformer3DModel.from_pretrained("rhymes-ai/Allegro", subfolder="transformer", dtype=torch.bfloat16).to("cuda")

AllegroTransformer3DModel[[diffusers.AllegroTransformer3DModel]]

diffusers.AllegroTransformer3DModel[[diffusers.AllegroTransformer3DModel]]

diffusers.AllegroTransformer3DModel(patch_size: int = 2, patch_size_t: int = 1, num_attention_heads: int = 24, attention_head_dim: int = 96, in_channels: int = 4, out_channels: int = 4, num_layers: int = 32, dropout: float = 0.0, cross_attention_dim: int = 2304, attention_bias: bool = True, sample_height: int = 90, sample_width: int = 160, sample_frames: int = 22, activation_fn: str = 'gelu-approximate', norm_elementwise_affine: bool = False, norm_eps: float = 1e-06, caption_channels: int = 4096, interpolation_scale_h: float = 2.0, interpolation_scale_w: float = 2.0, interpolation_scale_t: float = 2.2)

Source

forward[[diffusers.AllegroTransformer3DModel.forward]]

forward(hidden_states: Tensor, encoder_hidden_states: Tensor, timestep: LongTensor, attention_mask: typing.Optional[torch.Tensor] = None, encoder_attention_mask: typing.Optional[torch.Tensor] = None, image_rotary_emb: tuple[torch.Tensor, torch.Tensor] | None = None, return_dict: bool = True)

Source

Parameters:

hidden_states (torch.Tensor of shape (batch_size, num_channels, num_frames, height, width)) : Input hidden_states.

encoder_hidden_states (torch.Tensor of shape (batch_size, sequence_len, embed_dims)) : Conditional embeddings (embeddings computed from the input conditions such as prompts) to use.

timestep (torch.LongTensor) : Used to indicate denoising step.

attention_mask (torch.Tensor, optional) : Self-attention mask applied to hidden_states.

encoder_attention_mask (torch.Tensor, optional) : Cross-attention mask applied to encoder_hidden_states.

image_rotary_emb (tuple of torch.Tensor, optional) : Pre-computed rotary positional embeddings.

return_dict (bool, optional, defaults to True) : Whether or not to return a ~models.transformer_2d.Transformer2DModelOutput instead of a plain tuple.

Returns:

If return_dict is True, an ~models.transformer_2d.Transformer2DModelOutput is returned, otherwise a tuple where the first element is the sample tensor.

The AllegroTransformer3DModel forward method.

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

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

diffusers.models.modeling_outputs.Transformer2DModelOutput(sample: torch.Tensor)

Source

Parameters:

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.