# CosmosTransformer3DModel

A Diffusion Transformer model for 3D video-like data was introduced in [Cosmos World Foundation Model Platform for Physical AI](https://huggingface.co/papers/2501.03575) by NVIDIA.

The model can be loaded with the following code snippet.

```python
from diffusers import CosmosTransformer3DModel

transformer = CosmosTransformer3DModel.from_pretrained("nvidia/Cosmos-1.0-Diffusion-7B-Text2World", subfolder="transformer", torch_dtype=torch.bfloat16)
```

## CosmosTransformer3DModel[[diffusers.CosmosTransformer3DModel]]

#### diffusers.CosmosTransformer3DModel[[diffusers.CosmosTransformer3DModel]]

[Source](https://github.com/huggingface/diffusers/blob/v0.38.0/src/diffusers/models/transformers/transformer_cosmos.py#L554)

A Transformer model for video-like data used in [Cosmos](https://github.com/NVIDIA/Cosmos).

**Parameters:**

in_channels (`int`, defaults to `16`) : The number of channels in the input.

out_channels (`int`, defaults to `16`) : The number of channels in the output.

num_attention_heads (`int`, defaults to `32`) : The number of heads to use for multi-head attention.

attention_head_dim (`int`, defaults to `128`) : The number of channels in each attention head.

num_layers (`int`, defaults to `28`) : The number of layers of transformer blocks to use.

mlp_ratio (`float`, defaults to `4.0`) : The ratio of the hidden layer size to the input size in the feedforward network.

text_embed_dim (`int`, defaults to `4096`) : Input dimension of text embeddings from the text encoder.

adaln_lora_dim (`int`, defaults to `256`) : The hidden dimension of the Adaptive LayerNorm LoRA layer.

max_size (`tuple[int, int, int]`, defaults to `(128, 240, 240)`) : The maximum size of the input latent tensors in the temporal, height, and width dimensions.

patch_size (`tuple[int, int, int]`, defaults to `(1, 2, 2)`) : The patch size to use for patchifying the input latent tensors in the temporal, height, and width dimensions.

rope_scale (`tuple[float, float, float]`, defaults to `(2.0, 1.0, 1.0)`) : The scaling factor to use for RoPE in the temporal, height, and width dimensions.

concat_padding_mask (`bool`, defaults to `True`) : Whether to concatenate the padding mask to the input latent tensors.

extra_pos_embed_type (`str`, *optional*, defaults to `learnable`) : The type of extra positional embeddings to use. Can be one of `None` or `learnable`.

controlnet_block_every_n (`int`, *optional*) : Interval between transformer blocks that should receive control residuals (for example, `7` to inject after every seventh block). Required for Cosmos Transfer2.5.

img_context_dim_in (`int`, *optional*) : The dimension of the input image context feature vector, i.e. it is the D in [B, N, D].

img_context_num_tokens (`int`) : The number of tokens in the image context feature vector, i.e. it is the N in [B, N, D]. If `img_context_dim_in` is not provided, then this parameter is ignored.

img_context_dim_out (`int`) : The output dimension of the image context projection layer. If `img_context_dim_in` is not provided, then this parameter is ignored.

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

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

[Source](https://github.com/huggingface/diffusers/blob/v0.38.0/src/diffusers/models/modeling_outputs.py#L21)

The output of [Transformer2DModel](/docs/diffusers/v0.38.0/en/api/models/transformer2d#diffusers.Transformer2DModel).

**Parameters:**

sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` or `(batch size, num_vector_embeds - 1, num_latent_pixels)` if [Transformer2DModel](/docs/diffusers/v0.38.0/en/api/models/transformer2d#diffusers.Transformer2DModel) is discrete) : The hidden states output conditioned on the `encoder_hidden_states` input. If discrete, returns probability distributions for the unnoised latent pixels.