docs(generative_ai): Update Chat Completions API samples (GoogleCloudPlatform#13088)

* docs(generative_ai): Update Chat Completions API samples - Fix imports (tests were failing) - Add authentication sample - Combine credentials refresher region tags - Add samples for self-hosted models * Fix lint error * Update model endpoint id * Update generative_ai/chat_completions/chat_completions_authentication.py * Update generative_ai/chat_completions/chat_completions_authentication.py * Update generative_ai/chat_completions/chat_completions_authentication.py * Update Test to include new model Endpoint ID * Update Ednpoint ID * Change Model ID * Update endpoint to v1

Author: holtskinner

generative_ai/chat_completions/chat_completions_authentication.py

@@ -0,0 +1,50 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+def generate_text(project_id: str, location: str = "us-central1") -> object:
+ # [START generativeaionvertexai_gemini_chat_completions_authentication]
+ import openai
+
+ from google.auth import default
+ import google.auth.transport.requests
+
+ # TODO(developer): Update and un-comment below lines
+ # project_id = "PROJECT_ID"
+ # location = "us-central1"
+
+ # Programmatically get an access token
+ credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
+ credentials.refresh(google.auth.transport.requests.Request())
+ # Note: the credential lives for 1 hour by default (https://cloud.google.com/docs/authentication/token-types#at-lifetime); after expiration, it must be refreshed.
+
+ ##############################
+ # Choose one of the following:
+ ##############################
+
+ # If you are calling a Gemini model, set the ENDPOINT_ID variable to use openapi.
+ ENDPOINT_ID = "openapi"
+
+ # If you are calling a self-deployed model from Model Garden, set the
+ # ENDPOINT_ID variable and set the client's base URL to use your endpoint.
+ # ENDPOINT_ID = "YOUR_ENDPOINT_ID"
+
+ # OpenAI Client
+ client = openai.OpenAI(
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/{ENDPOINT_ID}",
+ api_key=credentials.token,
+ )
+ # [END generativeaionvertexai_gemini_chat_completions_authentication]
+
+ return client

generative_ai/chat_completions/chat_completions_credentials_refresher.py

@@ -15,7 +15,7 @@
 # Disable linting on `Any` type annotations (needed for OpenAI kwargs and attributes).
 # flake8: noqa ANN401
 
-# [START generativeaionvertexai_credentials_refresher_class]
+# [START generativeaionvertexai_credentials_refresher]
 from typing import Any
 
 import google.auth
@@ -25,16 +25,15 @@
 
 class OpenAICredentialsRefresher:
  def __init__(self, **kwargs: Any) -> None:
- # Set a dummy key here
- self.client = openai.OpenAI(**kwargs, api_key="DUMMY")
+ # Set a placeholder key here
+ self.client = openai.OpenAI(**kwargs, api_key="PLACEHOLDER")
  self.creds, self.project = google.auth.default(
  scopes=["https://www.googleapis.com/auth/cloud-platform"]
  )
 
  def __getattr__(self, name: str) -> Any:
  if not self.creds.valid:
- auth_req = google.auth.transport.requests.Request()
- self.creds.refresh(auth_req)
+ self.creds.refresh(google.auth.transport.requests.Request())
 
  if not self.creds.valid:
  raise RuntimeError("Unable to refresh auth")
@@ -43,18 +42,16 @@ def __getattr__(self, name: str) -> Any:
  return getattr(self.client, name)
 
 
-# [END generativeaionvertexai_credentials_refresher_class]
-
-
+# [END generativeaionvertexai_credentials_refresher]
 def generate_text(project_id: str, location: str = "us-central1") -> object:
- # [START generativeaionvertexai_credentials_refresher_usage]
+ # [START generativeaionvertexai_credentials_refresher]
 
  # TODO(developer): Update and un-comment below lines
  # project_id = "PROJECT_ID"
  # location = "us-central1"
 
  client = OpenAICredentialsRefresher(
- base_url=f"https://{location}-aiplatform.googleapis.com/v1beta1/projects/{project_id}/locations/{location}/endpoints/openapi",
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
  )
 
  response = client.chat.completions.create(
@@ -63,6 +60,6 @@ def generate_text(project_id: str, location: str = "us-central1") -> object:
  )
 
  print(response)
- # [END generativeaionvertexai_credentials_refresher_usage]
+ # [END generativeaionvertexai_credentials_refresher]
 
  return response

generative_ai/chat_completions/chat_completions_non_streaming_image.py

@@ -15,25 +15,23 @@
 
 def generate_text(project_id: str, location: str = "us-central1") -> object:
  # [START generativeaionvertexai_gemini_chat_completions_non_streaming_image]
- import vertexai
- import openai
 
- from google.auth import default, transport
+ from google.auth import default
+ import google.auth.transport.requests
+
+ import openai
 
  # TODO(developer): Update and un-comment below lines
  # project_id = "PROJECT_ID"
  # location = "us-central1"
 
- vertexai.init(project=project_id, location=location)
-
  # Programmatically get an access token
  credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
- auth_request = transport.requests.Request()
- credentials.refresh(auth_request)
+ credentials.refresh(google.auth.transport.requests.Request())
 
  # OpenAI Client
  client = openai.OpenAI(
- base_url=f"https://{location}-aiplatform.googleapis.com/v1beta1/projects/{project_id}/locations/{location}/endpoints/openapi",
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
  api_key=credentials.token,
  )
 

generative_ai/chat_completions/chat_completions_non_streaming_text.py

@@ -15,25 +15,22 @@
 
 def generate_text(project_id: str, location: str = "us-central1") -> object:
  # [START generativeaionvertexai_gemini_chat_completions_non_streaming]
- import vertexai
- import openai
+ from google.auth import default
+ import google.auth.transport.requests
 
- from google.auth import default, transport
+ import openai
 
  # TODO(developer): Update and un-comment below lines
  # project_id = "PROJECT_ID"
  # location = "us-central1"
 
- vertexai.init(project=project_id, location=location)
-
  # Programmatically get an access token
  credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
- auth_request = transport.requests.Request()
- credentials.refresh(auth_request)
+ credentials.refresh(google.auth.transport.requests.Request())
 
- # # OpenAI Client
+ # OpenAI Client
  client = openai.OpenAI(
- base_url=f"https://{location}-aiplatform.googleapis.com/v1beta1/projects/{project_id}/locations/{location}/endpoints/openapi",
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
  api_key=credentials.token,
  )
 

generative_ai/chat_completions/chat_completions_non_streaming_text_self_deployed.py

@@ -0,0 +1,52 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+def generate_text(
+ project_id: str,
+ location: str = "us-central1",
+ model_id: str = "gemma-2-9b-it",
+ endpoint_id: str = "YOUR_ENDPOINT_ID",
+) -> object:
+ # [START generativeaionvertexai_gemini_chat_completions_non_streaming_self_deployed]
+ from google.auth import default
+ import google.auth.transport.requests
+
+ import openai
+
+ # TODO(developer): Update and un-comment below lines
+ # project_id = "PROJECT_ID"
+ # location = "us-central1"
+ # model_id = "gemma-2-9b-it"
+ # endpoint_id = "YOUR_ENDPOINT_ID"
+
+ # Programmatically get an access token
+ credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
+ credentials.refresh(google.auth.transport.requests.Request())
+
+ # OpenAI Client
+ client = openai.OpenAI(
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/{endpoint_id}",
+ api_key=credentials.token,
+ )
+
+ response = client.chat.completions.create(
+ model=model_id,
+ messages=[{"role": "user", "content": "Why is the sky blue?"}],
+ )
+ print(response)
+
+ # [END generativeaionvertexai_gemini_chat_completions_non_streaming_self_deployed]
+
+ return response

generative_ai/chat_completions/chat_completions_streaming_image.py

@@ -15,25 +15,22 @@
 
 def generate_text(project_id: str, location: str = "us-central1") -> object:
  # [START generativeaionvertexai_gemini_chat_completions_streaming_image]
- import vertexai
- import openai
+ from google.auth import default
+ import google.auth.transport.requests
 
- from google.auth import default, transport
+ import openai
 
  # TODO(developer): Update and un-comment below lines
  # project_id = "PROJECT_ID"
  # location = "us-central1"
 
- vertexai.init(project=project_id, location=location)
-
  # Programmatically get an access token
  credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
- auth_request = transport.requests.Request()
- credentials.refresh(auth_request)
+ credentials.refresh(google.auth.transport.requests.Request())
 
  # OpenAI Client
  client = openai.OpenAI(
- base_url=f"https://{location}-aiplatform.googleapis.com/v1beta1/projects/{project_id}/locations/{location}/endpoints/openapi",
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
  api_key=credentials.token,
  )
 

generative_ai/chat_completions/chat_completions_streaming_text.py

@@ -15,25 +15,22 @@
 
 def generate_text(project_id: str, location: str = "us-central1") -> object:
  # [START generativeaionvertexai_gemini_chat_completions_streaming]
- import vertexai
- import openai
+ from google.auth import default
+ import google.auth.transport.requests
 
- from google.auth import default, transport
+ import openai
 
  # TODO(developer): Update and un-comment below lines
  # project_id = "PROJECT_ID"
  # location = "us-central1"
 
- vertexai.init(project=project_id, location=location)
-
  # Programmatically get an access token
  credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
- auth_request = transport.requests.Request()
- credentials.refresh(auth_request)
+ credentials.refresh(google.auth.transport.requests.Request())
 
  # OpenAI Client
  client = openai.OpenAI(
- base_url=f"https://{location}-aiplatform.googleapis.com/v1beta1/projects/{project_id}/locations/{location}/endpoints/openapi",
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/openapi",
  api_key=credentials.token,
  )
 

generative_ai/chat_completions/chat_completions_streaming_text_self_deployed.py

@@ -0,0 +1,54 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+def generate_text(
+ project_id: str,
+ location: str = "us-central1",
+ model_id: str = "gemma-2-9b-it",
+ endpoint_id: str = "YOUR_ENDPOINT_ID",
+) -> object:
+ # [START generativeaionvertexai_gemini_chat_completions_streaming_self_deployed]
+ from google.auth import default
+ import google.auth.transport.requests
+
+ import openai
+
+ # TODO(developer): Update and un-comment below lines
+ # project_id = "PROJECT_ID"
+ # location = "us-central1"
+ # model_id = "gemma-2-9b-it"
+ # endpoint_id = "YOUR_ENDPOINT_ID"
+
+ # Programmatically get an access token
+ credentials, _ = default(scopes=["https://www.googleapis.com/auth/cloud-platform"])
+ credentials.refresh(google.auth.transport.requests.Request())
+
+ # OpenAI Client
+ client = openai.OpenAI(
+ base_url=f"https://{location}-aiplatform.googleapis.com/v1/projects/{project_id}/locations/{location}/endpoints/{endpoint_id}",
+ api_key=credentials.token,
+ )
+
+ response = client.chat.completions.create(
+ model=model_id,
+ messages=[{"role": "user", "content": "Why is the sky blue?"}],
+ stream=True,
+ )
+ for chunk in response:
+ print(chunk)
+
+ # [END generativeaionvertexai_gemini_chat_completions_streaming_self_deployed]
+
+ return response

generative_ai/chat_completions/chat_completions_test.py

@@ -14,15 +14,25 @@
 
 import os
 
+import chat_completions_authentication
 import chat_completions_credentials_refresher
 import chat_completions_non_streaming_image
 import chat_completions_non_streaming_text
+import chat_completions_non_streaming_text_self_deployed
 import chat_completions_streaming_image
 import chat_completions_streaming_text
+import chat_completions_streaming_text_self_deployed
 
 
 PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
 LOCATION = "us-central1"
+SELF_HOSTED_MODEL_ID = "google/gemma-2-9b-it"
+ENDPOINT_ID = "6714120476014149632"
+
+
+def test_authentication() -> None:
+ response = chat_completions_authentication.generate_text(PROJECT_ID, LOCATION)
+ assert response
 
 
 def test_streaming_text() -> None:
@@ -50,3 +60,17 @@ def test_credentials_refresher() -> None:
  PROJECT_ID, LOCATION
  )
  assert response
+
+
+def test_streaming_text_self_deployed() -> None:
+ response = chat_completions_streaming_text_self_deployed.generate_text(
+ PROJECT_ID, LOCATION, SELF_HOSTED_MODEL_ID, ENDPOINT_ID
+ )
+ assert response
+
+
+def test_non_streaming_text_self_deployed() -> None:
+ response = chat_completions_non_streaming_text_self_deployed.generate_text(
+ PROJECT_ID, LOCATION, SELF_HOSTED_MODEL_ID, ENDPOINT_ID
+ )
+ assert response

generative_ai/chat_completions/requirements-test.txt

@@ -1,4 +1,4 @@
 backoff==2.2.1
-google-api-core==2.19.0
+google-api-core==2.24.0
 pytest==8.2.0
 pytest-asyncio==0.23.6