Bria fibo (#12545)

* Bria FIBO pipeline

* style fixs

* fix CR

* Refactor BriaFibo classes and update pipeline parameters

- Updated BriaFiboAttnProcessor and BriaFiboAttention classes to reflect changes from Flux equivalents.
- Modified the _unpack_latents method in BriaFiboPipeline to improve clarity.
- Increased the default max_sequence_length to 3000 and added a new optional parameter do_patching.
- Cleaned up test_pipeline_bria_fibo.py by removing unused imports and skipping unsupported tests.

* edit the docs of FIBO

* Remove unused BriaFibo imports and update CPU offload method in BriaFiboPipeline

* Refactor FIBO classes to BriaFibo naming convention

- Updated class names from FIBO to BriaFibo for consistency across the module.
- Modified instances of FIBOEmbedND, FIBOTimesteps, TextProjection, and TimestepProjEmbeddings to reflect the new naming.
- Ensured all references in the BriaFiboTransformer2DModel are updated accordingly.

* Add BriaFiboTransformer2DModel import to transformers module

* Remove unused BriaFibo imports from modular pipelines and add BriaFiboTransformer2DModel and BriaFiboPipeline classes to dummy objects for enhanced compatibility with torch and transformers.

* Update BriaFibo classes with copied documentation and fix import typo in pipeline module

- Added documentation comments indicating the source of copied code in BriaFiboTransformerBlock and _pack_latents methods.
- Corrected the import statement for BriaFiboPipeline in the pipelines module.

* Remove unused BriaFibo imports from __init__.py to streamline modular pipelines.

* Refactor documentation comments in BriaFibo classes to indicate inspiration from existing implementations

- Updated comments in BriaFiboAttnProcessor, BriaFiboAttention, and BriaFiboPipeline to reflect that the code is inspired by other modules rather than copied.
- Enhanced clarity on the origins of the methods to maintain proper attribution.

* change Inspired by to Based on

* add reference link and fix trailing whitespace

* Add BriaFiboTransformer2DModel documentation and update comments in BriaFibo classes

- Introduced a new documentation file for BriaFiboTransformer2DModel.
- Updated comments in BriaFiboAttnProcessor, BriaFiboAttention, and BriaFiboPipeline to clarify the origins of the code, indicating copied sources for better attribution.

---------
Co-authored-by: sayakpaul <spsayakpaul@gmail.com>

docs/source/en/_toctree.yml

@@ -323,6 +323,8 @@
         title: AllegroTransformer3DModel
       - local: api/models/aura_flow_transformer2d
         title: AuraFlowTransformer2DModel
+      - local: api/models/transformer_bria_fibo
+        title: BriaFiboTransformer2DModel
       - local: api/models/bria_transformer
         title: BriaTransformer2DModel
       - local: api/models/chroma_transformer
@@ -469,6 +471,8 @@
         title: BLIP-Diffusion
       - local: api/pipelines/bria_3_2
         title: Bria 3.2
+      - local: api/pipelines/bria_fibo
+        title: Bria Fibo
       - local: api/pipelines/chroma
         title: Chroma
       - local: api/pipelines/cogview3

docs/source/en/api/models/transformer_bria_fibo.md

+<!--Copyright 2025 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# BriaFiboTransformer2DModel
+
+A modified flux Transformer model from [Bria](https://huggingface.co/briaai/FIBO)
+
+## BriaFiboTransformer2DModel
+
+[[autodoc]] BriaFiboTransformer2DModel

docs/source/en/api/pipelines/bria_fibo.md

+<!--Copyright 2025 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# Bria Fibo
+
+Text-to-image models have mastered imagination - but not control. FIBO changes that.
+
+FIBO is trained on structured JSON captions up to 1,000+ words and designed to understand and control different visual parameters such as lighting, composition, color, and camera settings, enabling precise and reproducible outputs.
+
+With only 8 billion parameters, FIBO provides a new level of image quality, prompt adherence and proffesional control.
+
+FIBO is trained exclusively on a structured prompt and will not work with freeform text prompts.
+you can use the [FIBO-VLM-prompt-to-JSON](https://huggingface.co/briaai/FIBO-VLM-prompt-to-JSON) model or the [FIBO-gemini-prompt-to-JSON](https://huggingface.co/briaai/FIBO-gemini-prompt-to-JSON)  to convert your freeform text prompt to a structured JSON prompt.
+
+its not recommended to use freeform text prompts directly with FIBO, as it will not produce the best results.
+
+you can learn more about FIBO in  [Bria Fibo Hugging Face page](https://huggingface.co/briaai/FIBO).
+
+
+## Usage
+
+_As the model is gated, before using it with diffusers you first need to go to the [Bria Fibo Hugging Face page](https://huggingface.co/briaai/FIBO), fill in the form and accept the gate. Once you are in, you need to login so that your system knows you’ve accepted the gate._
+
+Use the command below to log in:
+
+```bash
+hf auth login
+```
+
+
+## BriaPipeline
+
+[[autodoc]] BriaPipeline
+	- all
+	- __call__
+

src/diffusers/__init__.py

@@ -198,6 +198,7 @@ else:
             "AutoencoderOobleck",
             "AutoencoderTiny",
             "AutoModel",
+            "BriaFiboTransformer2DModel",
             "BriaTransformer2DModel",
             "CacheMixin",
             "ChromaTransformer2DModel",
@@ -430,6 +431,7 @@ else:
             "AuraFlowPipeline",
             "BlipDiffusionControlNetPipeline",
             "BlipDiffusionPipeline",
+            "BriaFiboPipeline",
             "BriaPipeline",
             "ChromaImg2ImgPipeline",
             "ChromaPipeline",
@@ -901,6 +903,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             AutoencoderOobleck,
             AutoencoderTiny,
             AutoModel,
+            BriaFiboTransformer2DModel,
             BriaTransformer2DModel,
             CacheMixin,
             ChromaTransformer2DModel,
@@ -1103,6 +1106,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
             AudioLDM2UNet2DConditionModel,
             AudioLDMPipeline,
             AuraFlowPipeline,
+            BriaFiboPipeline,
             BriaPipeline,
             ChromaImg2ImgPipeline,
             ChromaPipeline,

src/diffusers/models/__init__.py

@@ -84,6 +84,7 @@ if is_torch_available():
     _import_structure["transformers.transformer_2d"] = ["Transformer2DModel"]
     _import_structure["transformers.transformer_allegro"] = ["AllegroTransformer3DModel"]
     _import_structure["transformers.transformer_bria"] = ["BriaTransformer2DModel"]
+    _import_structure["transformers.transformer_bria_fibo"] = ["BriaFiboTransformer2DModel"]
     _import_structure["transformers.transformer_chroma"] = ["ChromaTransformer2DModel"]
     _import_structure["transformers.transformer_cogview3plus"] = ["CogView3PlusTransformer2DModel"]
     _import_structure["transformers.transformer_cogview4"] = ["CogView4Transformer2DModel"]
@@ -174,6 +175,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .transformers import (
             AllegroTransformer3DModel,
             AuraFlowTransformer2DModel,
+            BriaFiboTransformer2DModel,
             BriaTransformer2DModel,
             ChromaTransformer2DModel,
             CogVideoXTransformer3DModel,

src/diffusers/models/transformers/__init__.py

@@ -18,6 +18,7 @@ if is_torch_available():
     from .transformer_2d import Transformer2DModel
     from .transformer_allegro import AllegroTransformer3DModel
     from .transformer_bria import BriaTransformer2DModel
+    from .transformer_bria_fibo import BriaFiboTransformer2DModel
     from .transformer_chroma import ChromaTransformer2DModel
     from .transformer_cogview3plus import CogView3PlusTransformer2DModel
     from .transformer_cogview4 import CogView4Transformer2DModel

src/diffusers/pipelines/__init__.py

@@ -128,6 +128,7 @@ else:
         "AnimateDiffVideoToVideoControlNetPipeline",
     ]
     _import_structure["bria"] = ["BriaPipeline"]
+    _import_structure["bria_fibo"] = ["BriaFiboPipeline"]
     _import_structure["flux"] = [
         "FluxControlPipeline",
         "FluxControlInpaintPipeline",
@@ -562,6 +563,7 @@ if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
         from .aura_flow import AuraFlowPipeline
         from .blip_diffusion import BlipDiffusionPipeline
         from .bria import BriaPipeline
+        from .bria_fibo import BriaFiboPipeline
         from .chroma import ChromaImg2ImgPipeline, ChromaPipeline
         from .cogvideo import (
             CogVideoXFunControlPipeline,

src/diffusers/pipelines/bria_fibo/__init__.py

+from typing import TYPE_CHECKING
+
+from ...utils import (
+    DIFFUSERS_SLOW_IMPORT,
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    get_objects_from_module,
+    is_torch_available,
+    is_transformers_available,
+)
+
+
+_dummy_objects = {}
+_import_structure = {}
+
+
+try:
+    if not (is_transformers_available() and is_torch_available()):
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    from ...utils import dummy_torch_and_transformers_objects  # noqa F403
+
+    _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
+else:
+    _import_structure["pipeline_bria_fibo"] = ["BriaFiboPipeline"]
+
+if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
+    try:
+        if not (is_transformers_available() and is_torch_available()):
+            raise OptionalDependencyNotAvailable()
+
+    except OptionalDependencyNotAvailable:
+        from ...utils.dummy_torch_and_transformers_objects import *
+    else:
+        from .pipeline_bria_fibo import BriaFiboPipeline
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(
+        __name__,
+        globals()["__file__"],
+        _import_structure,
+        module_spec=__spec__,
+    )
+
+    for name, value in _dummy_objects.items():
+        setattr(sys.modules[__name__], name, value)

src/diffusers/pipelines/bria_fibo/pipeline_output.py

+from dataclasses import dataclass
+from typing import List, Union
+
+import numpy as np
+import PIL.Image
+
+from ...utils import BaseOutput
+
+
+@dataclass
+class BriaFiboPipelineOutput(BaseOutput):
+    """
+    Output class for BriaFibo pipelines.
+
+    Args:
+        images (`List[PIL.Image.Image]` or `np.ndarray`)
+            List of denoised PIL images of length `batch_size` or numpy array of shape `(batch_size, height, width,
+            num_channels)`. PIL images or numpy array present the denoised images of the diffusion pipeline.
+    """
+
+    images: Union[List[PIL.Image.Image], np.ndarray]

src/diffusers/utils/dummy_pt_objects.py

@@ -588,6 +588,21 @@ class AutoModel(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class BriaFiboTransformer2DModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class BriaTransformer2DModel(metaclass=DummyObject):
     _backends = ["torch"]
 

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -482,6 +482,21 @@ class AuraFlowPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class BriaFiboPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class BriaPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

tests/models/transformers/test_models_transformer_bria_fibo.py

+# coding=utf-8
+# Copyright 2025 HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+
+import torch
+
+from diffusers import BriaFiboTransformer2DModel
+
+from ...testing_utils import enable_full_determinism, torch_device
+from ..test_modeling_common import ModelTesterMixin
+
+
+enable_full_determinism()
+
+
+class BriaFiboTransformerTests(ModelTesterMixin, unittest.TestCase):
+    model_class = BriaFiboTransformer2DModel
+    main_input_name = "hidden_states"
+    # We override the items here because the transformer under consideration is small.
+    model_split_percents = [0.8, 0.7, 0.7]
+
+    # Skip setting testing with default: AttnProcessor
+    uses_custom_attn_processor = True
+
+    @property
+    def dummy_input(self):
+        batch_size = 1
+        num_latent_channels = 48
+        num_image_channels = 3
+        height = width = 16
+        sequence_length = 32
+        embedding_dim = 64
+
+        hidden_states = torch.randn((batch_size, height * width, num_latent_channels)).to(torch_device)
+        encoder_hidden_states = torch.randn((batch_size, sequence_length, embedding_dim)).to(torch_device)
+        text_ids = torch.randn((sequence_length, num_image_channels)).to(torch_device)
+        image_ids = torch.randn((height * width, num_image_channels)).to(torch_device)
+        timestep = torch.tensor([1.0]).to(torch_device).expand(batch_size)
+
+        return {
+            "hidden_states": hidden_states,
+            "encoder_hidden_states": encoder_hidden_states,
+            "img_ids": image_ids,
+            "txt_ids": text_ids,
+            "timestep": timestep,
+            "text_encoder_layers": [encoder_hidden_states[:, :, :32], encoder_hidden_states[:, :, :32]],
+        }
+
+    @property
+    def input_shape(self):
+        return (16, 16)
+
+    @property
+    def output_shape(self):
+        return (256, 48)
+
+    def prepare_init_args_and_inputs_for_common(self):
+        init_dict = {
+            "patch_size": 1,
+            "in_channels": 48,
+            "num_layers": 1,
+            "num_single_layers": 1,
+            "attention_head_dim": 8,
+            "num_attention_heads": 2,
+            "joint_attention_dim": 64,
+            "text_encoder_dim": 32,
+            "pooled_projection_dim": None,
+            "axes_dims_rope": [0, 4, 4],
+        }
+
+        inputs_dict = self.dummy_input
+        return init_dict, inputs_dict
+
+    def test_gradient_checkpointing_is_applied(self):
+        expected_set = {"BriaFiboTransformer2DModel"}
+        super().test_gradient_checkpointing_is_applied(expected_set=expected_set)

tests/pipelines/bria_fibo/test_pipeline_bria_fibo.py

+# Copyright 2024 Bria AI and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+
+import numpy as np
+import torch
+from transformers import AutoTokenizer
+from transformers.models.smollm3.modeling_smollm3 import SmolLM3Config, SmolLM3ForCausalLM
+
+from diffusers import (
+    AutoencoderKLWan,
+    BriaFiboPipeline,
+    FlowMatchEulerDiscreteScheduler,
+)
+from diffusers.models.transformers.transformer_bria_fibo import BriaFiboTransformer2DModel
+from tests.pipelines.test_pipelines_common import PipelineTesterMixin
+
+from ...testing_utils import (
+    enable_full_determinism,
+    torch_device,
+)
+
+
+enable_full_determinism()
+
+
+class BriaFiboPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
+    pipeline_class = BriaFiboPipeline
+    params = frozenset(["prompt", "height", "width", "guidance_scale"])
+    batch_params = frozenset(["prompt"])
+    test_xformers_attention = False
+    test_layerwise_casting = False
+    test_group_offloading = False
+    supports_dduf = False
+
+    def get_dummy_components(self):
+        torch.manual_seed(0)
+        transformer = BriaFiboTransformer2DModel(
+            patch_size=1,
+            in_channels=16,
+            num_layers=1,
+            num_single_layers=1,
+            attention_head_dim=8,
+            num_attention_heads=2,
+            joint_attention_dim=64,
+            text_encoder_dim=32,
+            pooled_projection_dim=None,
+            axes_dims_rope=[0, 4, 4],
+        )
+
+        torch.manual_seed(0)
+        vae = AutoencoderKLWan(
+            base_dim=160,
+            decoder_base_dim=256,
+            num_res_blocks=2,
+            out_channels=12,
+            patch_size=2,
+            scale_factor_spatial=16,
+            scale_factor_temporal=4,
+            temperal_downsample=[False, True, True],
+            z_dim=16,
+        )
+
+        scheduler = FlowMatchEulerDiscreteScheduler()
+
+        torch.manual_seed(0)
+        text_encoder = SmolLM3ForCausalLM(SmolLM3Config(hidden_size=32))
+        tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-t5")
+
+        components = {
+            "scheduler": scheduler,
+            "text_encoder": text_encoder,
+            "tokenizer": tokenizer,
+            "transformer": transformer,
+            "vae": vae,
+        }
+        return components
+
+    def get_dummy_inputs(self, device, seed=0):
+        if str(device).startswith("mps"):
+            generator = torch.manual_seed(seed)
+        else:
+            generator = torch.Generator(device="cpu").manual_seed(seed)
+
+        inputs = {
+            "prompt": "{'text': 'A painting of a squirrel eating a burger'}",
+            "negative_prompt": "bad, ugly",
+            "generator": generator,
+            "num_inference_steps": 2,
+            "guidance_scale": 5.0,
+            "height": 32,
+            "width": 32,
+            "output_type": "np",
+        }
+        return inputs
+
+    @unittest.skip(reason="will not be supported due to dim-fusion")
+    def test_encode_prompt_works_in_isolation(self):
+        pass
+
+    def test_bria_fibo_different_prompts(self):
+        pipe = self.pipeline_class(**self.get_dummy_components())
+        pipe = pipe.to(torch_device)
+        inputs = self.get_dummy_inputs(torch_device)
+        output_same_prompt = pipe(**inputs).images[0]
+
+        inputs = self.get_dummy_inputs(torch_device)
+        inputs["prompt"] = "a different prompt"
+        output_different_prompts = pipe(**inputs).images[0]
+
+        max_diff = np.abs(output_same_prompt - output_different_prompts).max()
+        assert max_diff > 1e-6
+
+    def test_image_output_shape(self):
+        pipe = self.pipeline_class(**self.get_dummy_components())
+        pipe = pipe.to(torch_device)
+        inputs = self.get_dummy_inputs(torch_device)
+
+        height_width_pairs = [(32, 32), (64, 64), (32, 64)]
+        for height, width in height_width_pairs:
+            expected_height = height
+            expected_width = width
+
+            inputs.update({"height": height, "width": width})
+            image = pipe(**inputs).images[0]
+            output_height, output_width, _ = image.shape
+            assert (output_height, output_width) == (expected_height, expected_width)