diffusers/pr_12229/en/api/models/mochi_transformer3d.md

MochiTransformer3DModel

A Diffusion Transformer model for 3D video-like data was introduced in Mochi-1 Preview by Genmo.

The model can be loaded with the following code snippet.

from diffusers import MochiTransformer3DModel

transformer = MochiTransformer3DModel.from_pretrained("genmo/mochi-1-preview", subfolder="transformer", torch_dtype=torch.float16).to("cuda")

MochiTransformer3DModel[[diffusers.MochiTransformer3DModel]]

class diffusers.MochiTransformer3DModeldiffusers.MochiTransformer3DModelhttps://github.com/huggingface/diffusers/blob/vr_12229/src/diffusers/models/transformers/transformer_mochi.py#L308[{"name": "patch_size", "val": ": int = 2"}, {"name": "num_attention_heads", "val": ": int = 24"}, {"name": "attention_head_dim", "val": ": int = 128"}, {"name": "num_layers", "val": ": int = 48"}, {"name": "pooled_projection_dim", "val": ": int = 1536"}, {"name": "in_channels", "val": ": int = 12"}, {"name": "out_channels", "val": ": typing.Optional[int] = None"}, {"name": "qk_norm", "val": ": str = 'rms_norm'"}, {"name": "text_embed_dim", "val": ": int = 4096"}, {"name": "time_embed_dim", "val": ": int = 256"}, {"name": "activation_fn", "val": ": str = 'swiglu'"}, {"name": "max_sequence_length", "val": ": int = 256"}]- patch_size (int, defaults to 2) -- The size of the patches to use in the patch embedding layer.

• num_attention_heads (int, defaults to 24) -- The number of heads to use for multi-head attention.
• attention_head_dim (int, defaults to 128) -- The number of channels in each head.
• num_layers (int, defaults to 48) -- The number of layers of Transformer blocks to use.
• in_channels (int, defaults to 12) -- The number of channels in the input.
• out_channels (int, optional, defaults to None) -- The number of channels in the output.
• qk_norm (str, defaults to "rms_norm") -- The normalization layer to use.
• text_embed_dim (int, defaults to 4096) -- Input dimension of text embeddings from the text encoder.
• time_embed_dim (int, defaults to 256) -- Output dimension of timestep embeddings.
• activation_fn (str, defaults to "swiglu") -- Activation function to use in feed-forward.
• max_sequence_length (int, defaults to 256) -- The maximum sequence length of text embeddings supported.0

A Transformer model for video-like data introduced in Mochi.

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

class diffusers.models.modeling_outputs.Transformer2DModelOutputdiffusers.models.modeling_outputs.Transformer2DModelOutputhttps://github.com/huggingface/diffusers/blob/vr_12229/src/diffusers/models/modeling_outputs.py#L20[{"name": "*args", "val": ""}, {"name": "**kwargs", "val": ""}]- 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.0

The output of Transformer2DModel.