Update OpenAI image query default model to GPT-4o

CHANGELOG.md

@@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Changed
 - Default the value of `azure_deployment` on all Azure Drivers to the model the Driver is using.
 - Field `azure_ad_token` on all Azure Drivers is no longer serializable.
+- Default standard OpenAI and Azure OpenAI image query model to `gpt-4o`.
+- **BREAKING**: Updated OpenAI-based image query drivers to remove Vision from the name.
 
 ## [0.25.1] - 2024-05-15
 

docs/griptape-framework/drivers/image-query-drivers.md

@@ -64,15 +64,15 @@ print(result)
 !!! info
     While the `max_tokens` field is optional, it is recommended to set this to a value that corresponds to the desired response length. Without an explicit value, the model will default to very short responses. See [OpenAI's documentation](https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them) for more information on how to relate token count to response length.
 
-The [OpenAiVisionImageQueryDriver](../../reference/griptape/drivers/image_query/openai_vision_image_query_driver.md) is used to query images using the OpenAI Vision API. Here is an example of how to use it:
+The [OpenAiVisionImageQueryDriver](../../reference/griptape/drivers/image_query/openai_image_query_driver.md) is used to query images using the OpenAI Vision API. Here is an example of how to use it:
 
 ```python
-from griptape.drivers import OpenAiVisionImageQueryDriver
+from griptape.drivers import OpenAiImageQueryDriver
 from griptape.engines import ImageQueryEngine
 from griptape.loaders import ImageLoader
 
-driver = OpenAiVisionImageQueryDriver(
-    model="gpt-4-vision-preview",
+driver = OpenAiImageQueryDriver(
+    model="gpt-4o",
     max_tokens=256,
 )
 
@@ -91,19 +91,19 @@ engine.run("Describe the weather in the image", [image_artifact])
 !!! info
     In order to use the `gpt-4-vision-preview` model on Azure OpenAI, the `gpt-4` model must be deployed with the version set to `vision-preview`. More information can be found in the [Azure OpenAI documentation](https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/gpt-with-vision).
 
-The [AzureOpenAiVisionImageQueryDriver](../../reference/griptape/drivers/image_query/azure_openai_vision_image_query_driver.md) is used to query images using the Azure OpenAI Vision API. Here is an example of how to use it:
+The [AzureOpenAiVisionImageQueryDriver](../../reference/griptape/drivers/image_query/azure_openai_image_query_driver.md) is used to query images using the Azure OpenAI Vision API. Here is an example of how to use it:
 
 ```python
 import os
-from griptape.drivers import AzureOpenAiVisionImageQueryDriver
+from griptape.drivers import AzureOpenAiImageQueryDriver
 from griptape.engines import ImageQueryEngine
 from griptape.loaders import ImageLoader
 
-driver = AzureOpenAiVisionImageQueryDriver(
-    azure_endpoint=os.environ["AZURE_OPENAI_ENDPOINT_3"],
-    api_key=os.environ["AZURE_OPENAI_API_KEY_3"],
-    model="gpt-4",
-    azure_deployment="gpt-4-vision-preview",
+driver = AzureOpenAiImageQueryDriver(
+    azure_endpoint=os.environ["AZURE_OPENAI_ENDPOINT_2"],
+    api_key=os.environ["AZURE_OPENAI_API_KEY_2"],
+    model="gpt-4o",
+    azure_deployment="gpt-4o",
     max_tokens=256,
 )
 

docs/griptape-framework/engines/image-query-engines.md

@@ -2,15 +2,15 @@
 
 The [Image Query Engine](../../reference/griptape/engines/image_query/image_query_engine.md) is used to execute natural language queries on the contents of images. You can specify the provider and model used to query the image by providing the Engine with a particular [Image Query Driver](../drivers/image-query-drivers.md).
 
-All Image Query Drivers default to a `max_tokens` of 256. You can tune this value based on your use case and the [Image Query Driver](../drivers/image-query-drivers.md) you are providing. 
+All Image Query Drivers default to a `max_tokens` of 256. You can tune this value based on your use case and the [Image Query Driver](../drivers/image-query-drivers.md) you are providing.
 
 ```python
-from griptape.drivers import OpenAiVisionImageQueryDriver
+from griptape.drivers import OpenAiImageQueryDriver
 from griptape.engines import ImageQueryEngine
 from griptape.loaders import ImageLoader
 
-driver = OpenAiVisionImageQueryDriver(
-    model="gpt-4-vision-preview",
+driver = OpenAiImageQueryDriver(
+    model="gpt-4o",
     max_tokens=256
 )
 

docs/griptape-framework/structures/config.md

@@ -38,7 +38,7 @@ agent = Agent(
         api_key=os.environ["AZURE_OPENAI_API_KEY_3"]
     ).merge_config({
         "image_query_driver": {
-            "azure_deployment": "gpt-4-vision-preview",
+            "azure_deployment": "gpt-4o",
         },
     }),
 )

docs/griptape-framework/structures/tasks.md

@@ -633,15 +633,14 @@ This Task accepts two inputs: a query (represented by either a string or a [Text
 
 ```python
 from griptape.engines import ImageQueryEngine
-from griptape.drivers import OpenAiVisionImageQueryDriver
+from griptape.drivers import OpenAiImageQueryDriver
 from griptape.tasks import ImageQueryTask
 from griptape.loaders import ImageLoader
 from griptape.structures import Pipeline
 
-
 # Create a driver configured to use OpenAI's GPT-4 Vision model.
-driver = OpenAiVisionImageQueryDriver(
-    model="gpt-4-vision-preview",
+driver = OpenAiImageQueryDriver(
+    model="gpt-4o",
     max_tokens=100,
 )
 

docs/griptape-tools/official-tools/image-query-client.md

@@ -5,14 +5,14 @@ This tool allows Agents to execute natural language queries on the contents of i
 ```python
 from griptape.structures import Agent
 from griptape.tools import ImageQueryClient
-from griptape.drivers import OpenAiVisionImageQueryDriver
-from griptape.engines import ImageQueryEngine 
+from griptape.drivers import OpenAiImageQueryDriver
+from griptape.engines import ImageQueryEngine
 
 # Create an Image Query Driver.
-driver = OpenAiVisionImageQueryDriver(
-    model="gpt-4-vision-preview"
+driver = OpenAiImageQueryDriver(
+    model="gpt-4o"
 )
-    
+
 # Create an Image Query Engine configured to use the driver.
 engine = ImageQueryEngine(
     image_query_driver=driver,

griptape/config/azure_openai_structure_config.py

@@ -7,7 +7,7 @@
     AzureOpenAiChatPromptDriver,
     AzureOpenAiEmbeddingDriver,
     AzureOpenAiImageGenerationDriver,
-    AzureOpenAiVisionImageQueryDriver,
+    AzureOpenAiImageQueryDriver,
     BasePromptDriver,
     BaseEmbeddingDriver,
     BaseImageGenerationDriver,
@@ -41,7 +41,7 @@ class AzureOpenAiStructureConfig(StructureConfig):
     prompt_driver: BasePromptDriver = field(
         default=Factory(
             lambda self: AzureOpenAiChatPromptDriver(
-                model="gpt-4",
+                model="gpt-4o",
                 azure_endpoint=self.azure_endpoint,
                 api_key=self.api_key,
                 azure_ad_token=self.azure_ad_token,
@@ -69,8 +69,8 @@ class AzureOpenAiStructureConfig(StructureConfig):
     )
     image_query_driver: BaseImageQueryDriver = field(
         default=Factory(
-            lambda self: AzureOpenAiVisionImageQueryDriver(
-                model="gpt-4",
+            lambda self: AzureOpenAiImageQueryDriver(
+                model="gpt-4o",
                 azure_endpoint=self.azure_endpoint,
                 api_key=self.api_key,
                 azure_ad_token=self.azure_ad_token,

griptape/config/openai_structure_config.py

@@ -11,7 +11,7 @@
     OpenAiChatPromptDriver,
     OpenAiEmbeddingDriver,
     OpenAiImageGenerationDriver,
-    OpenAiVisionImageQueryDriver,
+    OpenAiImageQueryDriver,
 )
 
 
@@ -26,9 +26,7 @@ class OpenAiStructureConfig(StructureConfig):
         metadata={"serializable": True},
     )
     image_query_driver: BaseImageQueryDriver = field(
-        default=Factory(lambda: OpenAiVisionImageQueryDriver(model="gpt-4-vision-preview")),
-        kw_only=True,
-        metadata={"serializable": True},
+        default=Factory(lambda: OpenAiImageQueryDriver(model="gpt-4o")), kw_only=True, metadata={"serializable": True}
     )
     embedding_driver: BaseEmbeddingDriver = field(
         default=Factory(lambda: OpenAiEmbeddingDriver(model="text-embedding-3-small")),

griptape/drivers/__init__.py

@@ -79,9 +79,9 @@
 from .image_query.base_image_query_driver import BaseImageQueryDriver
 from .image_query.base_multi_model_image_query_driver import BaseMultiModelImageQueryDriver
 from .image_query.dummy_image_query_driver import DummyImageQueryDriver
-from .image_query.openai_vision_image_query_driver import OpenAiVisionImageQueryDriver
+from .image_query.openai_image_query_driver import OpenAiImageQueryDriver
 from .image_query.anthropic_image_query_driver import AnthropicImageQueryDriver
-from .image_query.azure_openai_vision_image_query_driver import AzureOpenAiVisionImageQueryDriver
+from .image_query.azure_openai_image_query_driver import AzureOpenAiImageQueryDriver
 from .image_query.amazon_bedrock_image_query_driver import AmazonBedrockImageQueryDriver
 
 from .web_scraper.base_web_scraper_driver import BaseWebScraperDriver
@@ -175,8 +175,8 @@
     "BaseImageQueryModelDriver",
     "BedrockClaudeImageQueryModelDriver",
     "BaseImageQueryDriver",
-    "OpenAiVisionImageQueryDriver",
-    "AzureOpenAiVisionImageQueryDriver",
+    "OpenAiImageQueryDriver",
+    "AzureOpenAiImageQueryDriver",
     "DummyImageQueryDriver",
     "AnthropicImageQueryDriver",
     "BaseMultiModelImageQueryDriver",

griptape/drivers/image_query/azure_openai_vision_image_query_driver.py → griptape/drivers/image_query/azure_openai_image_query_driver.py

@@ -4,11 +4,11 @@
 
 from attr import define, field, Factory
 import openai
-from griptape.drivers.image_query.openai_vision_image_query_driver import OpenAiVisionImageQueryDriver
+from griptape.drivers.image_query.openai_image_query_driver import OpenAiImageQueryDriver
 
 
 @define
-class AzureOpenAiVisionImageQueryDriver(OpenAiVisionImageQueryDriver):
+class AzureOpenAiImageQueryDriver(OpenAiImageQueryDriver):
     """Driver for Azure-hosted OpenAI image query API.
 
     Attributes:

griptape/drivers/image_query/openai_vision_image_query_driver.py → griptape/drivers/image_query/openai_image_query_driver.py

@@ -16,7 +16,7 @@
 
 
 @define
-class OpenAiVisionImageQueryDriver(BaseImageQueryDriver):
+class OpenAiImageQueryDriver(BaseImageQueryDriver):
     model: str = field(kw_only=True, metadata={"serializable": True})
     api_type: str = field(default=openai.api_type, kw_only=True)
     api_version: Optional[str] = field(default=openai.api_version, kw_only=True, metadata={"serializable": True})

tests/unit/config/test_azure_openai_structure_config.py

@@ -22,8 +22,8 @@ def test_to_dict(self, config):
             "prompt_driver": {
                 "type": "AzureOpenAiChatPromptDriver",
                 "base_url": None,
-                "model": "gpt-4",
-                "azure_deployment": "gpt-4",
+                "model": "gpt-4o",
+                "azure_deployment": "gpt-4o",
                 "azure_endpoint": "http://localhost:8080",
                 "api_version": "2023-05-15",
                 "organization": None,
@@ -62,12 +62,12 @@ def test_to_dict(self, config):
                 "base_url": None,
                 "image_quality": "auto",
                 "max_tokens": 256,
-                "model": "gpt-4",
+                "model": "gpt-4o",
                 "api_version": "2024-02-01",
-                "azure_deployment": "gpt-4",
+                "azure_deployment": "gpt-4o",
                 "azure_endpoint": "http://localhost:8080",
                 "organization": None,
-                "type": "AzureOpenAiVisionImageQueryDriver",
+                "type": "AzureOpenAiImageQueryDriver",
             },
             "vector_store_driver": {
                 "embedding_driver": {

tests/unit/config/test_openai_structure_config.py

@@ -50,9 +50,9 @@ def test_to_dict(self, config):
                 "base_url": None,
                 "image_quality": "auto",
                 "max_tokens": 256,
-                "model": "gpt-4-vision-preview",
+                "model": "gpt-4o",
                 "organization": None,
-                "type": "OpenAiVisionImageQueryDriver",
+                "type": "OpenAiImageQueryDriver",
             },
             "vector_store_driver": {
                 "embedding_driver": {

tests/unit/drivers/image_query/test_azure_openai_image_query_driver.py

@@ -1,6 +1,6 @@
 import pytest
 from unittest.mock import Mock
-from griptape.drivers import AzureOpenAiVisionImageQueryDriver
+from griptape.drivers import AzureOpenAiImageQueryDriver
 from griptape.artifacts import ImageArtifact
 
 
@@ -13,15 +13,13 @@ def mock_completion_create(self, mocker):
         return mock_chat_create
 
     def test_init(self):
-        assert AzureOpenAiVisionImageQueryDriver(
+        assert AzureOpenAiImageQueryDriver(
             azure_endpoint="test-endpoint", azure_deployment="test-deployment", model="gpt-4"
         )
-        assert (
-            AzureOpenAiVisionImageQueryDriver(azure_endpoint="test-endpoint", model="gpt-4").azure_deployment == "gpt-4"
-        )
+        assert AzureOpenAiImageQueryDriver(azure_endpoint="test-endpoint", model="gpt-4").azure_deployment == "gpt-4"
 
     def test_try_query_defaults(self, mock_completion_create):
-        driver = AzureOpenAiVisionImageQueryDriver(
+        driver = AzureOpenAiImageQueryDriver(
             azure_endpoint="test-endpoint", azure_deployment="test-deployment", model="gpt-4"
         )
         test_prompt_string = "Prompt String"
@@ -36,7 +34,7 @@ def test_try_query_defaults(self, mock_completion_create):
         assert text_artifact.value == "expected_output_text"
 
     def test_try_query_max_tokens(self, mock_completion_create):
-        driver = AzureOpenAiVisionImageQueryDriver(
+        driver = AzureOpenAiImageQueryDriver(
             azure_endpoint="test-endpoint", azure_deployment="test-deployment", model="gpt-4", max_tokens=1024
         )
         test_prompt_string = "Prompt String"
@@ -50,7 +48,7 @@ def test_try_query_max_tokens(self, mock_completion_create):
 
     def test_try_query_multiple_choices(self, mock_completion_create):
         mock_completion_create.return_value.choices.append(Mock(message=Mock(content="expected_output_text2")))
-        driver = AzureOpenAiVisionImageQueryDriver(
+        driver = AzureOpenAiImageQueryDriver(
             azure_endpoint="test-endpoint", azure_deployment="test-deployment", model="gpt-4"
         )
 

tests/unit/drivers/image_query/test_openai_image_query_driver.py

@@ -1,6 +1,6 @@
 import pytest
 from unittest.mock import Mock
-from griptape.drivers import OpenAiVisionImageQueryDriver
+from griptape.drivers import OpenAiImageQueryDriver
 from griptape.artifacts import ImageArtifact
 
 
@@ -13,10 +13,10 @@ def mock_completion_create(self, mocker):
         return mock_chat_create
 
     def test_init(self):
-        assert OpenAiVisionImageQueryDriver(model="gpt-4-vision-preview")
+        assert OpenAiImageQueryDriver(model="gpt-4-vision-preview")
 
     def test_try_query_defaults(self, mock_completion_create):
-        driver = OpenAiVisionImageQueryDriver(model="gpt-4-vision-preview")
+        driver = OpenAiImageQueryDriver(model="gpt-4-vision-preview")
         test_prompt_string = "Prompt String"
         test_binary_data = b"test-data"
         test_image = ImageArtifact(value=test_binary_data, width=100, height=100, format="png")
@@ -29,7 +29,7 @@ def test_try_query_defaults(self, mock_completion_create):
         assert text_artifact.value == "expected_output_text"
 
     def test_try_query_max_tokens(self, mock_completion_create):
-        driver = OpenAiVisionImageQueryDriver(model="gpt-4-vision-preview", max_tokens=1024)
+        driver = OpenAiImageQueryDriver(model="gpt-4-vision-preview", max_tokens=1024)
         test_prompt_string = "Prompt String"
         test_binary_data = b"test-data"
         test_image = ImageArtifact(value=test_binary_data, width=100, height=100, format="png")
@@ -41,7 +41,7 @@ def test_try_query_max_tokens(self, mock_completion_create):
 
     def test_try_query_multiple_choices(self, mock_completion_create):
         mock_completion_create.return_value.choices.append(Mock(message=Mock(content="expected_output_text2")))
-        driver = OpenAiVisionImageQueryDriver(model="gpt-4-vision-preview")
+        driver = OpenAiImageQueryDriver(model="gpt-4-vision-preview")
 
         with pytest.raises(Exception):
             driver.try_query("Prompt String", [ImageArtifact(value=b"test-data", width=100, height=100, format="png")])