fix(bedrock.py): fix response format for bedrock image generation response

Fixes https://github.com/BerriAI/litellm/issues/5010

litellm/llms/bedrock.py

@@ -13,6 +13,7 @@ from enum import Enum
 from typing import Any, Callable, List, Optional, Union
 
 import httpx
+from openai.types.image import Image
 
 import litellm
 from litellm.litellm_core_utils.core_helpers import map_finish_reason
@@ -1413,10 +1414,10 @@ def embedding(
 def image_generation(
     model: str,
     prompt: str,
+    model_response: ImageResponse,
+    optional_params: dict,
     timeout=None,
     logging_obj=None,
-    model_response=None,
-    optional_params=None,
     aimg_generation=False,
 ):
     """
@@ -1513,9 +1514,10 @@ def image_generation(
     if model_response is None:
         model_response = ImageResponse()
 
-    image_list: List = []
+    image_list: List[Image] = []
     for artifact in response_body["artifacts"]:
-        image_dict = {"url": artifact["base64"]}
+        _image = Image(b64_json=artifact["base64"])
+        image_list.append(_image)
 
-    model_response.data = image_dict
+    model_response.data = image_list
     return model_response

litellm/tests/test_image_generation.py

@@ -158,7 +158,11 @@ def test_image_generation_bedrock():
             model="bedrock/stability.stable-diffusion-xl-v1",
             aws_region_name="us-west-2",
         )
+
         print(f"response: {response}")
+        from openai.types.images_response import ImagesResponse
+
+        ImagesResponse.model_validate(response.model_dump())
     except litellm.RateLimitError as e:
         pass
     except litellm.ContentPolicyViolationError:

litellm/types/utils.py

@@ -950,16 +950,18 @@ class ImageObject(OpenAIObject):
             return self.dict()
 
 
-class ImageResponse(OpenAIObject):
+from openai.types.images_response import ImagesResponse as OpenAIImageResponse
-    created: Optional[int] = None
 
-    data: Optional[List[ImageObject]] = None
-
-    usage: Optional[dict] = None
 
+class ImageResponse(OpenAIImageResponse):
     _hidden_params: dict = {}
 
-    def __init__(self, created=None, data=None, response_ms=None):
+    def __init__(
+        self,
+        created: Optional[int] = None,
+        data: Optional[list] = None,
+        response_ms=None,
+    ):
         if response_ms:
             _response_ms = response_ms
         else:
@@ -967,14 +969,16 @@ class ImageResponse(OpenAIObject):
         if data:
             data = data
         else:
-            data = None
+            data = []
 
         if created:
             created = created
         else:
-            created = None
+            created = int(time.time())
 
-        super().__init__(data=data, created=created)
+        _data = {"data": data, "created": created}
+
+        super().__init__(**_data)
         self.usage = {"prompt_tokens": 0, "completion_tokens": 0, "total_tokens": 0}
 
     def __contains__(self, key):