huggingface · manueldeprada · Sep 1, 2025 · Aug 26, 2025 · Aug 26, 2025 · Aug 26, 2025
diff --git a/docs/source/ar/notebooks.md b/docs/source/ar/notebooks.md
@@ -39,7 +39,6 @@
 | [كيفية ضبط نموذج بدقة على التلخيص](https://github.com/huggingface/notebooks/blob/main/examples/summarization.ipynb)| يوضح كيفية معالجة البيانات مسبقًا وضبط نموذج مُدرَّب مسبقًا بدقة على XSUM. | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)|
 | [كيفية تدريب نموذج لغة من البداية](https://github.com/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| تسليط الضوء على جميع الخطوات لتدريب نموذج Transformer بشكل فعال على بيانات مخصصة | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)|
 | [كيفية إنشاء نص](https://github.com/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| كيفية استخدام أساليب فك التشفير المختلفة لإنشاء اللغة باستخدام المحولات | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)|
-| [كيفية إنشاء نص (مع قيود)](https://github.com/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| كيفية توجيه إنشاء اللغة باستخدام القيود التي يوفرها المستخدم | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)|
 | [Reformer](https://github.com/huggingface/blog/blob/main/notebooks/03_reformer.ipynb)| كيف يدفع Reformer حدود النمذجة اللغوية | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)|
 
 #### رؤية الكمبيوتر[[pytorch-cv]]

diff --git a/docs/source/ja/internal/generation_utils.md b/docs/source/ja/internal/generation_utils.md
@@ -300,28 +300,6 @@ generation_output[:2]
 [[autodoc]] MaxTimeCriteria
     - __call__
 
-## Constraints
-
-[`Constraint`] を使用すると、生成時に出力に特定のトークンまたはシーケンスが含まれるように強制できます。これは PyTorch 実装でのみ利用可能であることに注意してください。
-
-[[autodoc]] Constraint
-
-[[autodoc]] PhrasalConstraint
-
-[[autodoc]] DisjunctiveConstraint
-
-[[autodoc]] ConstraintListState
-
-## BeamSearch
-
-[[autodoc]] BeamScorer
-    - process
-    - finalize
-
-[[autodoc]] ConstrainedBeamSearchScorer
-    - process
-    - finalize
-
 ## Streamers
 
 [[autodoc]] TextStreamer

diff --git a/docs/source/ko/internal/generation_utils.md b/docs/source/ko/internal/generation_utils.md
@@ -305,28 +305,6 @@ generation_output[:2]
 [[autodoc]] EosTokenCriteria
     - __call__
 
-## Constraint [[transformers.Constraint]]
-
-[`Constraint`]는 생성 출력에 특정 토큰이나 시퀀스를 강제로 포함시키는 데 사용됩니다. 이 기능은 PyTorch 구현에만 제공됩니다.
-
-[[autodoc]] Constraint
-
-[[autodoc]] PhrasalConstraint
-
-[[autodoc]] DisjunctiveConstraint
-
-[[autodoc]] ConstraintListState
-
-## 빔 검색 (BeamSearch) [[transformers.BeamScorer]]
-
-[[autodoc]] BeamScorer
-    - process
-    - finalize
-
-[[autodoc]] ConstrainedBeamSearchScorer
-    - process
-    - finalize
-
 ## 스트리머 (Streamers) [[transformers.TextStreamer]]
 
 [[autodoc]] TextStreamer

diff --git a/docs/source/zh/internal/generation_utils.md b/docs/source/zh/internal/generation_utils.md
@@ -295,28 +295,6 @@ generation_output[:2]
 [[autodoc]] MaxTimeCriteria
     - __call__
 
-## Constraints
-
-可以使用[`Constraint`]来强制生成结果包含输出中的特定tokens或序列。请注意，这仅适用于我们的PyTorch实现。
-
-[[autodoc]] Constraint
-
-[[autodoc]] PhrasalConstraint
-
-[[autodoc]] DisjunctiveConstraint
-
-[[autodoc]] ConstraintListState
-
-## BeamSearch
-
-[[autodoc]] BeamScorer
-    - process
-    - finalize
-
-[[autodoc]] ConstrainedBeamSearchScorer
-    - process
-    - finalize
-
 ## Streamers
 
 [[autodoc]] TextStreamer

diff --git a/notebooks/README.md b/notebooks/README.md
@@ -56,7 +56,6 @@ You can open any page of the documentation as a notebook in Colab (there is a bu
 | [How to fine-tune a model on summarization](https://github.com/huggingface/notebooks/blob/main/examples/summarization.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on XSUM. | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)|
 | [How to train a language model from scratch](https://github.com/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| Highlight all the steps to effectively train Transformer model on custom data | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)|
 | [How to generate text](https://github.com/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| How to use different decoding methods for language generation with transformers | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)|
-| [How to generate text (with constraints)](https://github.com/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| How to guide language generation with user-provided constraints | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)|
 | [Reformer](https://github.com/huggingface/blog/blob/main/notebooks/03_reformer.ipynb)| How Reformer pushes the limits of language modeling | [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)| [![Open in AWS Studio](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)|
 
 #### Computer Vision[[pytorch-cv]]

diff --git a/src/transformers/generation/beam_constraints.py b/src/transformers/generation/beam_constraints.py
@@ -1,7 +1,13 @@
 from abc import ABC, abstractmethod
 from typing import Optional
 
+from ..utils import logging
 
+
+logger = logging.get_logger(__name__)
+
+
+# TODO joao, manuel: remove in v4.58.0
 class Constraint(ABC):
     r"""Abstract base class for all constraints that can be applied during generation.
     It must define how the constraint can be satisfied.
@@ -18,6 +24,9 @@ class Constraint(ABC):
     """
 
     def __init__(self):
+        logger.warning_once(
+            "Importing `Constraint` classes is deprecated and will be removed in v4.58.0. Constrained beam search has been moved to the Hub: https://hf.co/transformers-community/constrained-beam-search. Please import using `from transformers.generation import Constraint` instead."
+        )
         # test for the above condition
         self.test()