feat: Release API key support for GenerateContent to Public Preview

PiperOrigin-RevId: 685926683

google/cloud/aiplatform/initializer.py

@@ -221,6 +221,14 @@ def init(
                     f"{api_transport} is not a valid transport type. "
                     + f"Valid transport types: {VALID_TRANSPORT_TYPES}"
                 )
+        else:
+            # Raise error if api_transport other than rest is specified for usage with API key.
+            if not project and not api_transport:
+                api_transport = "rest"
+            elif not project and api_transport != "rest":
+                raise ValueError(
+                    f"{api_transport} is not supported with API keys. "
+                )
         if location:
             utils.validate_region(location)
         if experiment_description and experiment is None:
@@ -236,6 +244,9 @@ def init(
                 logging.info("project/location updated, reset Experiment config.")
             metadata._experiment_tracker.reset()
 
+        if project and api_key:
+            logging.info("Both a project and API key have been provided. The project will take precedence over the API key.")
+
         # Then we change the main state
         if api_endpoint is not None:
             self._api_endpoint = api_endpoint
@@ -438,7 +449,13 @@ def get_client_options(
 
         api_endpoint = self.api_endpoint
 
+        if api_endpoint is None and not self._project and not self._location and not location_override:
+            # Default endpoint is location invariant if using API key
+            api_endpoint = "aiplatform.googleapis.com"
+
+        # If both project and API key are passed in, project takes precedence.
         if api_endpoint is None:
+            # Form the default endpoint to use with no API key.
             if not (self.location or location_override):
                 raise ValueError(
                     "No location found. Provide or initialize SDK with a location."

tests/unit/vertex_rag/test_rag_data.py

@@ -306,7 +306,7 @@ class TestRagDataManagement:
     def setup_method(self):
         importlib.reload(aiplatform.initializer)
         importlib.reload(aiplatform)
-        aiplatform.init()
+        aiplatform.init(project=tc.TEST_PROJECT, location=tc.TEST_REGION)
 
     def teardown_method(self):
         aiplatform.initializer.global_pool.shutdown(wait=True)

vertexai/generative_models/_generative_models.py

@@ -388,13 +388,25 @@ def __init__(
     def _prediction_client(self) -> prediction_service.PredictionServiceClient:
         # Switch to @functools.cached_property once its available.
         if not getattr(self, "_prediction_client_value", None):
-            self._prediction_client_value = (
-                aiplatform_initializer.global_config.create_client(
-                    client_class=prediction_service.PredictionServiceClient,
-                    location_override=self._location,
-                    prediction_client=True,
+            if (
+                aiplatform_initializer.global_config.api_key
+                and not aiplatform_initializer.global_config.project
+            ):
+                self._prediction_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=prediction_service.PredictionServiceClient,
+                        api_key=aiplatform_initializer.global_config.api_key,
+                        prediction_client=True,
+                    )
+                )
+            else:
+                self._prediction_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=prediction_service.PredictionServiceClient,
+                        location_override=self._location,
+                        prediction_client=True,
+                    )
                 )
-            )
         return self._prediction_client_value
 
     @property
@@ -403,26 +415,46 @@ def _prediction_async_client(
     ) -> prediction_service.PredictionServiceAsyncClient:
         # Switch to @functools.cached_property once its available.
         if not getattr(self, "_prediction_async_client_value", None):
-            self._prediction_async_client_value = (
-                aiplatform_initializer.global_config.create_client(
-                    client_class=prediction_service.PredictionServiceAsyncClient,
-                    location_override=self._location,
-                    prediction_client=True,
+            if (
+                aiplatform_initializer.global_config.api_key
+                and not aiplatform_initializer.global_config.project
+            ):
+                raise RuntimeError(
+                    "Using an api key is not supported yet for async clients."
+                )
+            else:
+                self._prediction_async_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=prediction_service.PredictionServiceAsyncClient,
+                        location_override=self._location,
+                        prediction_client=True,
+                    )
                 )
-            )
         return self._prediction_async_client_value
 
     @property
     def _llm_utility_client(self) -> llm_utility_service.LlmUtilityServiceClient:
         # Switch to @functools.cached_property once its available.
         if not getattr(self, "_llm_utility_client_value", None):
-            self._llm_utility_client_value = (
-                aiplatform_initializer.global_config.create_client(
-                    client_class=llm_utility_service.LlmUtilityServiceClient,
-                    location_override=self._location,
-                    prediction_client=True,
+            if (
+                aiplatform_initializer.global_config.api_key
+                and not aiplatform_initializer.global_config.project
+            ):
+                self._llm_utility_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=llm_utility_service.LlmUtilityServiceClient,
+                        api_key=aiplatform_initializer.global_config.api_key,
+                        prediction_client=True,
+                    )
+                )
+            else:
+                self._llm_utility_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=llm_utility_service.LlmUtilityServiceClient,
+                        location_override=self._location,
+                        prediction_client=True,
+                    )
                 )
-            )
         return self._llm_utility_client_value
 
     @property
@@ -431,13 +463,21 @@ def _llm_utility_async_client(
     ) -> llm_utility_service.LlmUtilityServiceAsyncClient:
         # Switch to @functools.cached_property once its available.
         if not getattr(self, "_llm_utility_async_client_value", None):
-            self._llm_utility_async_client_value = (
-                aiplatform_initializer.global_config.create_client(
-                    client_class=llm_utility_service.LlmUtilityServiceAsyncClient,
-                    location_override=self._location,
-                    prediction_client=True,
+            if (
+                aiplatform_initializer.global_config.api_key
+                and not aiplatform_initializer.global_config.project
+            ):
+                raise RuntimeError(
+                    "Using an api key is not supported yet for async clients."
+                )
+            else:
+                self._llm_utility_async_client_value = (
+                    aiplatform_initializer.global_config.create_client(
+                        client_class=llm_utility_service.LlmUtilityServiceAsyncClient,
+                        location_override=self._location,
+                        prediction_client=True,
+                    )
                 )
-            )
         return self._llm_utility_async_client_value
 
     def _prepare_request(