migrate to transformers v5

.github/workflows/pr_tests.yml

@@ -92,8 +92,9 @@ jobs:
             runner: aws-general-8-plus
             image: diffusers/diffusers-pytorch-cpu
             report: torch_example_cpu
+        transformers_version: ["main"]
 
-    name: ${{ matrix.config.name }}
+    name: ${{ matrix.config.name }} (transformers ${{ matrix.transformers_version }})
 
     runs-on:
       group: ${{ matrix.config.runner }}
@@ -155,7 +156,7 @@ jobs:
       if: ${{ always() }}
       uses: actions/upload-artifact@v6
       with:
-        name: pr_${{ matrix.config.framework }}_${{ matrix.config.report }}_test_reports
+        name: pr_${{ matrix.config.framework }}_${{ matrix.config.report }}_transformers_${{ matrix.transformers_version }}_test_reports
         path: reports
 
   run_staging_tests:
@@ -220,8 +221,10 @@ jobs:
     needs: [check_code_quality, check_repository_consistency]
     strategy:
       fail-fast: false
+      matrix:
+        transformers_version: ["main"]
 
-    name: LoRA tests with PEFT main
+    name: LoRA tests with PEFT main (transformers ${{ matrix.transformers_version }})
 
     runs-on:
       group: aws-general-8-plus
@@ -247,9 +250,12 @@ jobs:
         uv pip install -U peft@git+https://github.com/huggingface/peft.git --no-deps
         uv pip install -U tokenizers
         uv pip uninstall accelerate && uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git --no-deps
-        #uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
-        uv pip uninstall transformers huggingface_hub && uv pip install transformers==4.57.1
-
+        if [ "${{ matrix.transformers_version }}" = "main" ]; then
+          uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
+        else
+          uv pip uninstall transformers huggingface_hub && uv pip install transformers==${{ matrix.transformers_version }}
+        fi
+
     - name: Environment
       run: |
         python utils/print_env.py
@@ -275,6 +281,6 @@ jobs:
       if: ${{ always() }}
       uses: actions/upload-artifact@v6
       with:
-        name: pr_main_test_reports
+        name: pr_lora_transformers_${{ matrix.transformers_version }}_test_reports
         path: reports
 

.github/workflows/pr_tests_gpu.yml

@@ -14,6 +14,7 @@ on:
       - "tests/pipelines/test_pipelines_common.py"
       - "tests/models/test_modeling_common.py"
       - "examples/**/*.py"
+      - ".github/**.yml"
   workflow_dispatch:
 
 concurrency:
@@ -106,13 +107,14 @@ jobs:
           path: reports
 
   torch_pipelines_cuda_tests:
-    name: Torch Pipelines CUDA Tests
+    name: Torch Pipelines CUDA Tests (transformers ${{ matrix.transformers_version }})
     needs: setup_torch_cuda_pipeline_matrix
     strategy:
       fail-fast: false
       max-parallel: 8
       matrix:
         module: ${{ fromJson(needs.setup_torch_cuda_pipeline_matrix.outputs.pipeline_test_matrix) }}
+        transformers_version: ["main"]
     runs-on:
       group: aws-g4dn-2xlarge
     container:
@@ -131,8 +133,12 @@ jobs:
         run: |
           uv pip install -e ".[quality]"
           uv pip uninstall accelerate && uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
-          #uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
-          uv pip uninstall transformers huggingface_hub && uv pip install transformers==4.57.1
+          if [ "${{ matrix.transformers_version }}" = "main" ]; then
+            uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
+          else
+            uv pip uninstall transformers huggingface_hub && uv pip install transformers==${{ matrix.transformers_version }}
+          fi
+
 
       - name: Environment
         run: |
@@ -172,11 +178,11 @@ jobs:
         if: ${{ always() }}
         uses: actions/upload-artifact@v6
         with:
-          name: pipeline_${{ matrix.module }}_test_reports
+          name: pipeline_${{ matrix.module }}_transformers_${{ matrix.transformers_version }}_test_reports
           path: reports
 
   torch_cuda_tests:
-    name: Torch CUDA Tests
+    name: Torch CUDA Tests (transformers ${{ matrix.transformers_version }})
     needs: [check_code_quality, check_repository_consistency]
     runs-on:
       group: aws-g4dn-2xlarge
@@ -191,6 +197,7 @@ jobs:
       max-parallel: 4
       matrix:
         module: [models, schedulers, lora, others]
+        transformers_version: ["main"]
     steps:
     - name: Checkout diffusers
       uses: actions/checkout@v6
@@ -202,8 +209,12 @@ jobs:
         uv pip install -e ".[quality]"
         uv pip install peft@git+https://github.com/huggingface/peft.git
         uv pip uninstall accelerate && uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
-        #uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
-        uv pip uninstall transformers huggingface_hub && uv pip install transformers==4.57.1
+        if [ "${{ matrix.transformers_version }}" = "main" ]; then
+          uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
+        else
+          uv pip uninstall transformers huggingface_hub && uv pip install transformers==${{ matrix.transformers_version }}
+        fi
+
 
     - name: Environment
       run: |
@@ -241,12 +252,16 @@ jobs:
       if: ${{ always() }}
       uses: actions/upload-artifact@v6
       with:
-        name: torch_cuda_test_reports_${{ matrix.module }}
+        name: torch_cuda_test_reports_${{ matrix.module }}_transformers_${{ matrix.transformers_version }}
         path: reports
 
   run_examples_tests:
-    name: Examples PyTorch CUDA tests on Ubuntu
+    name: Examples PyTorch CUDA tests on Ubuntu (transformers ${{ matrix.transformers_version }})
     needs: [check_code_quality, check_repository_consistency]
+    strategy:
+      fail-fast: false
+      matrix:
+        transformers_version: ["main"]
     runs-on:
       group: aws-g4dn-2xlarge
 
@@ -264,8 +279,11 @@ jobs:
         nvidia-smi
     - name: Install dependencies
       run: |
-        #uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
-        uv pip uninstall transformers huggingface_hub && uv pip install transformers==4.57.1
+        if [ "${{ matrix.transformers_version }}" = "main" ]; then
+          uv pip uninstall transformers huggingface_hub && uv pip install --prerelease allow -U transformers@git+https://github.com/huggingface/transformers.git
+        else
+          uv pip uninstall transformers huggingface_hub && uv pip install transformers==${{ matrix.transformers_version }}
+        fi
         uv pip install -e ".[quality,training]"
 
     - name: Environment
@@ -289,6 +307,6 @@ jobs:
       if: ${{ always() }}
       uses: actions/upload-artifact@v6
       with:
-        name: examples_test_reports
+        name: examples_transformers_${{ matrix.transformers_version }}_test_reports
         path: reports
 

examples/custom_diffusion/test_custom_diffusion.py

@@ -17,6 +17,9 @@
 import os
 import sys
 import tempfile
+import unittest
+
+from diffusers.utils import is_transformers_version
 
 
 sys.path.append("..")
@@ -30,6 +33,7 @@
 logger.addHandler(stream_handler)
 
 
+@unittest.skipIf(is_transformers_version(">=", "4.57.5"), "Size mismatch")
 class CustomDiffusion(ExamplesTestsAccelerate):
     def test_custom_diffusion(self):
         with tempfile.TemporaryDirectory() as tmpdir:

src/diffusers/hooks/_common.py

@@ -48,6 +48,7 @@
     torch.nn.ConvTranspose2d,
     torch.nn.ConvTranspose3d,
     torch.nn.Linear,
+    torch.nn.Embedding,
     # TODO(aryan): look into torch.nn.LayerNorm, torch.nn.GroupNorm later, seems to be causing some issues with CogVideoX
     # because of double invocation of the same norm layer in CogVideoXLayerNorm
 )

src/diffusers/loaders/textual_inversion.py

@@ -22,7 +22,12 @@
 from torch import nn
 
 from ..models.modeling_utils import load_state_dict
-from ..utils import _get_model_file, is_accelerate_available, is_transformers_available, logging
+from ..utils import (
+    _get_model_file,
+    is_accelerate_available,
+    is_transformers_available,
+    logging,
+)
 
 
 if is_transformers_available():

src/diffusers/pipelines/audioldm2/pipeline_audioldm2.py

@@ -502,6 +502,10 @@ def encode_prompt(
                         text_input_ids,
                         attention_mask=attention_mask,
                     )
+                    # Extract the pooler output if it's a BaseModelOutputWithPooling (Transformers v5+)
+                    # otherwise use it directly (Transformers v4)
+                    if hasattr(prompt_embeds, "pooler_output"):
+                        prompt_embeds = prompt_embeds.pooler_output
                     # append the seq-len dim: (bs, hidden_size) -> (bs, seq_len, hidden_size)
                     prompt_embeds = prompt_embeds[:, None, :]
                     # make sure that we attend to this single hidden-state
@@ -610,6 +614,10 @@ def encode_prompt(
                         uncond_input_ids,
                         attention_mask=negative_attention_mask,
                     )
+                    # Extract the pooler output if it's a BaseModelOutputWithPooling (Transformers v5+)
+                    # otherwise use it directly (Transformers v4)
+                    if hasattr(negative_prompt_embeds, "pooler_output"):
+                        negative_prompt_embeds = negative_prompt_embeds.pooler_output
                     # append the seq-len dim: (bs, hidden_size) -> (bs, seq_len, hidden_size)
                     negative_prompt_embeds = negative_prompt_embeds[:, None, :]
                     # make sure that we attend to this single hidden-state

src/diffusers/pipelines/cosmos/pipeline_cosmos2_5_predict.py

@@ -287,6 +287,9 @@ def _get_prompt_embeds(
                 truncation=True,
                 padding="max_length",
             )
+            input_ids = (
+                input_ids["input_ids"] if not isinstance(input_ids, list) and "input_ids" in input_ids else input_ids
+            )
             input_ids = torch.LongTensor(input_ids)
             input_ids_batch.append(input_ids)
 

src/diffusers/pipelines/cosmos/pipeline_cosmos2_5_transfer.py

@@ -262,6 +262,9 @@ def _get_prompt_embeds(
                 truncation=True,
                 padding="max_length",
             )
+            input_ids = (
+                input_ids["input_ids"] if not isinstance(input_ids, list) and "input_ids" in input_ids else input_ids
+            )
             input_ids = torch.LongTensor(input_ids)
             input_ids_batch.append(input_ids)
 

src/diffusers/pipelines/kandinsky/text_encoder.py

@@ -20,6 +20,8 @@ def __init__(self, config, *args, **kwargs):
         self.LinearTransformation = torch.nn.Linear(
             in_features=config.transformerDimensions, out_features=config.numDims
         )
+        if hasattr(self, "post_init"):
+            self.post_init()
 
     def forward(self, input_ids, attention_mask):
         embs = self.transformer(input_ids=input_ids, attention_mask=attention_mask)[0]

src/diffusers/pipelines/kolors/text_encoder.py

@@ -781,6 +781,9 @@ def __init__(self, config: ChatGLMConfig, device=None, empty_init=True):
             self.prefix_encoder = PrefixEncoder(config)
             self.dropout = torch.nn.Dropout(0.1)
 
+        if hasattr(self, "post_init"):
+            self.post_init()
+
     def get_input_embeddings(self):
         return self.embedding.word_embeddings
 
@@ -810,7 +813,7 @@ def forward(
         output_hidden_states = (
             output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
         )
-        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        use_cache = use_cache if use_cache is not None else getattr(self.config, "use_cache", None)
         return_dict = return_dict if return_dict is not None else self.config.use_return_dict
 
         batch_size, seq_length = input_ids.shape

src/diffusers/pipelines/pipeline_utils.py

@@ -341,6 +341,7 @@ def is_saveable_module(name, value):
             save_method_accept_safe = "safe_serialization" in save_method_signature.parameters
             save_method_accept_variant = "variant" in save_method_signature.parameters
             save_method_accept_max_shard_size = "max_shard_size" in save_method_signature.parameters
+            save_method_accept_peft_format = "save_peft_format" in save_method_signature.parameters
 
             save_kwargs = {}
             if save_method_accept_safe:
@@ -350,6 +351,11 @@ def is_saveable_module(name, value):
             if save_method_accept_max_shard_size and max_shard_size is not None:
                 # max_shard_size is expected to not be None in ModelMixin
                 save_kwargs["max_shard_size"] = max_shard_size
+            if save_method_accept_peft_format:
+                # Set save_peft_format=False for transformers>=5.0.0 compatibility
+                # In transformers 5.0.0+, the default save_peft_format=True adds "base_model.model" prefix
+                # to adapter keys, but from_pretrained expects keys without this prefix
+                save_kwargs["save_peft_format"] = False
 
             save_method(os.path.join(save_directory, pipeline_component_name), **save_kwargs)
 

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2.py

@@ -17,7 +17,7 @@
 
 import regex as re
 import torch
-from transformers import AutoTokenizer, UMT5EncoderModel
+from transformers import AutoTokenizer, T5EncoderModel, UMT5EncoderModel
 
 from ...callbacks import MultiPipelineCallbacks, PipelineCallback
 from ...loaders import SkyReelsV2LoraLoaderMixin
@@ -132,7 +132,7 @@ class SkyReelsV2Pipeline(DiffusionPipeline, SkyReelsV2LoraLoaderMixin):
     def __init__(
         self,
         tokenizer: AutoTokenizer,
-        text_encoder: UMT5EncoderModel,
+        text_encoder: T5EncoderModel | UMT5EncoderModel,
         transformer: SkyReelsV2Transformer3DModel,
         vae: AutoencoderKLWan,
         scheduler: UniPCMultistepScheduler,

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing.py

@@ -19,7 +19,7 @@
 from typing import Any, Callable
 
 import torch
-from transformers import AutoTokenizer, UMT5EncoderModel
+from transformers import AutoTokenizer, T5EncoderModel, UMT5EncoderModel
 
 from ...callbacks import MultiPipelineCallbacks, PipelineCallback
 from ...loaders import SkyReelsV2LoraLoaderMixin
@@ -153,7 +153,7 @@ class SkyReelsV2DiffusionForcingPipeline(DiffusionPipeline, SkyReelsV2LoraLoader
     def __init__(
         self,
         tokenizer: AutoTokenizer,
-        text_encoder: UMT5EncoderModel,
+        text_encoder: T5EncoderModel | UMT5EncoderModel,
         transformer: SkyReelsV2Transformer3DModel,
         vae: AutoencoderKLWan,
         scheduler: UniPCMultistepScheduler,

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_i2v.py

@@ -20,7 +20,7 @@
 
 import PIL
 import torch
-from transformers import AutoTokenizer, UMT5EncoderModel
+from transformers import AutoTokenizer, T5EncoderModel, UMT5EncoderModel
 
 from diffusers.image_processor import PipelineImageInput
 from diffusers.utils.torch_utils import randn_tensor
@@ -158,7 +158,7 @@ class SkyReelsV2DiffusionForcingImageToVideoPipeline(DiffusionPipeline, SkyReels
     def __init__(
         self,
         tokenizer: AutoTokenizer,
-        text_encoder: UMT5EncoderModel,
+        text_encoder: T5EncoderModel | UMT5EncoderModel,
         transformer: SkyReelsV2Transformer3DModel,
         vae: AutoencoderKLWan,
         scheduler: UniPCMultistepScheduler,

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_v2v.py

@@ -21,7 +21,7 @@
 
 import torch
 from PIL import Image
-from transformers import AutoTokenizer, UMT5EncoderModel
+from transformers import AutoTokenizer, T5EncoderModel, UMT5EncoderModel
 
 from ...callbacks import MultiPipelineCallbacks, PipelineCallback
 from ...loaders import SkyReelsV2LoraLoaderMixin
@@ -214,7 +214,7 @@ class SkyReelsV2DiffusionForcingVideoToVideoPipeline(DiffusionPipeline, SkyReels
     def __init__(
         self,
         tokenizer: AutoTokenizer,
-        text_encoder: UMT5EncoderModel,
+        text_encoder: T5EncoderModel | UMT5EncoderModel,
         transformer: SkyReelsV2Transformer3DModel,
         vae: AutoencoderKLWan,
         scheduler: UniPCMultistepScheduler,

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_i2v.py

@@ -18,7 +18,7 @@
 import PIL
 import regex as re
 import torch
-from transformers import AutoTokenizer, CLIPProcessor, CLIPVisionModelWithProjection, UMT5EncoderModel
+from transformers import AutoTokenizer, CLIPProcessor, CLIPVisionModelWithProjection, T5EncoderModel, UMT5EncoderModel
 
 from ...callbacks import MultiPipelineCallbacks, PipelineCallback
 from ...image_processor import PipelineImageInput
@@ -157,7 +157,7 @@ class SkyReelsV2ImageToVideoPipeline(DiffusionPipeline, SkyReelsV2LoraLoaderMixi
     def __init__(
         self,
         tokenizer: AutoTokenizer,
-        text_encoder: UMT5EncoderModel,
+        text_encoder: T5EncoderModel | UMT5EncoderModel,
         image_encoder: CLIPVisionModelWithProjection,
         image_processor: CLIPProcessor,
         transformer: SkyReelsV2Transformer3DModel,