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

BriaTransformer2DModel

A modified flux Transformer model from Bria

BriaTransformer2DModel[[diffusers.BriaTransformer2DModel]]

class diffusers.BriaTransformer2DModeldiffusers.BriaTransformer2DModelhttps://github.com/huggingface/diffusers/blob/vr_12229/src/diffusers/models/transformers/transformer_bria.py#L506[{"name": "patch_size", "val": ": int = 1"}, {"name": "in_channels", "val": ": int = 64"}, {"name": "num_layers", "val": ": int = 19"}, {"name": "num_single_layers", "val": ": int = 38"}, {"name": "attention_head_dim", "val": ": int = 128"}, {"name": "num_attention_heads", "val": ": int = 24"}, {"name": "joint_attention_dim", "val": ": int = 4096"}, {"name": "pooled_projection_dim", "val": ": int = None"}, {"name": "guidance_embeds", "val": ": bool = False"}, {"name": "axes_dims_rope", "val": ": typing.List[int] = [16, 56, 56]"}, {"name": "rope_theta", "val": " = 10000"}, {"name": "time_theta", "val": " = 10000"}]- patch_size (int) -- Patch size to turn the input data into small patches.

• in_channels (int, optional, defaults to 16) -- The number of channels in the input.
• num_layers (int, optional, defaults to 18) -- The number of layers of MMDiT blocks to use.
• num_single_layers (int, optional, defaults to 18) -- The number of layers of single DiT blocks to use.
• attention_head_dim (int, optional, defaults to 64) -- The number of channels in each head.
• num_attention_heads (int, optional, defaults to 18) -- The number of heads to use for multi-head attention.
• joint_attention_dim (int, optional) -- The number of encoder_hidden_states dimensions to use.
• pooled_projection_dim (int) -- Number of dimensions to use when projecting the pooled_projections.
• guidance_embeds (bool, defaults to False) -- Whether to use guidance embeddings.0

The Transformer model introduced in Flux. Based on FluxPipeline with several changes:

• no pooled embeddings
• We use zero padding for prompts
• No guidance embedding since this is not a distilled version Reference: https://blackforestlabs.ai/announcing-black-forest-labs/

forwarddiffusers.BriaTransformer2DModel.forwardhttps://github.com/huggingface/diffusers/blob/vr_12229/src/diffusers/models/transformers/transformer_bria.py#L584[{"name": "hidden_states", "val": ": Tensor"}, {"name": "encoder_hidden_states", "val": ": Tensor = None"}, {"name": "pooled_projections", "val": ": Tensor = None"}, {"name": "timestep", "val": ": LongTensor = None"}, {"name": "img_ids", "val": ": Tensor = None"}, {"name": "txt_ids", "val": ": Tensor = None"}, {"name": "guidance", "val": ": Tensor = None"}, {"name": "attention_kwargs", "val": ": typing.Optional[typing.Dict[str, typing.Any]] = None"}, {"name": "return_dict", "val": ": bool = True"}, {"name": "controlnet_block_samples", "val": " = None"}, {"name": "controlnet_single_block_samples", "val": " = None"}]- hidden_states (torch.FloatTensor of shape (batch size, channel, height, width)) -- Input hidden_states.

• encoder_hidden_states (torch.FloatTensor of shape (batch size, sequence_len, embed_dims)) -- Conditional embeddings (embeddings computed from the input conditions such as prompts) to use.
• pooled_projections (torch.FloatTensor of shape (batch_size, projection_dim)) -- Embeddings projected from the embeddings of input conditions.
• timestep ( torch.LongTensor) -- Used to indicate denoising step.
• block_controlnet_hidden_states -- (list of torch.Tensor): A list of tensors that if specified are added to the residuals of transformer blocks.
• attention_kwargs (dict, optional) -- A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor.
• return_dict (bool, optional, defaults to True) -- Whether or not to return a ~models.transformer_2d.Transformer2DModelOutput instead of a plain tuple.0If return_dict is True, an ~models.transformer_2d.Transformer2DModelOutput is returned, otherwise a tuple where the first element is the sample tensor.

The BriaTransformer2DModel forward method.