diff --git a/CHANGELOG.md b/CHANGELOG.md
index 19c3ceae..3ae9c17f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog],
 and this project adheres to [Semantic Versioning].
 
+## [Unreleased]
+
+### Added
+
+- Add `HFEndpointRetriever` to use a Hf Space as a model provider
+
 ## [v5.0.0] - 2025-02-19
 
 ### Added
diff --git a/src/vidore_benchmark/retrievers/cohere_api_retriever.py b/src/vidore_benchmark/retrievers/cohere_api_retriever.py
index 6d0263e8..1e4064e6 100644
--- a/src/vidore_benchmark/retrievers/cohere_api_retriever.py
+++ b/src/vidore_benchmark/retrievers/cohere_api_retriever.py
@@ -20,6 +20,16 @@
 
 @register_vision_retriever("cohere")
 class CohereAPIRetriever(BaseVisionRetriever):
+    """
+    Retriever that uses the Cohere API to generate embeddings.
+
+    Example usage:
+
+    ```python
+    retriever = CohereAPIRetriever(pretrained_model_name_or_path="embed-english-v3.0")
+    ```
+    """
+
     def __init__(
         self,
         pretrained_model_name_or_path: str = "embed-english-v3.0",
diff --git a/src/vidore_benchmark/retrievers/hf_api_retriever.py b/src/vidore_benchmark/retrievers/hf_api_retriever.py
index e872bcd0..0b1a0a8c 100644
--- a/src/vidore_benchmark/retrievers/hf_api_retriever.py
+++ b/src/vidore_benchmark/retrievers/hf_api_retriever.py
@@ -21,6 +21,17 @@
 
 @register_vision_retriever("hf-endpoint")
 class HFEndpointRetriever(BaseVisionRetriever):
+    """
+    Retriever that uses a custom Hugging Face endpoint that serves an arbritrary model to generate embeddings.
+    The `pretrained_model_name_or_path` should be the URL of the endpoint.
+
+    Example usage:
+
+    ```python
+    retriever = HFEndpointRetriever(pretrained_model_name_or_path="https://your_endpoint_url.endpoints.huggingface.cloud")
+    ```
+    """
+
     def __init__(self, pretrained_model_name_or_path: str, **kwargs):
         super().__init__(use_visual_embedding=True)
         self.url = pretrained_model_name_or_path