chore(genai): Update Gemini Model IDs

genai/batch_prediction/batchpredict_with_bq.py

@@ -18,7 +18,7 @@ def generate_content(output_uri: str) -> str:
     import time
 
     from google import genai
-    from google.genai.types import CreateBatchJobConfig, JobState, HttpOptions
+    from google.genai.types import CreateBatchJobConfig, HttpOptions, JobState
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
 
@@ -28,7 +28,7 @@ def generate_content(output_uri: str) -> str:
     job = client.batches.create(
         # To use a tuned model, set the model param to your tuned model using the following format:
         # model="projects/{PROJECT_ID}/locations/{LOCATION}/models/{MODEL_ID}
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         src="bq://storage-samples.generative_ai.batch_requests_for_multimodal_input",
         config=CreateBatchJobConfig(dest=output_uri),
     )

genai/batch_prediction/batchpredict_with_gcs.py

@@ -18,7 +18,7 @@ def generate_content(output_uri: str) -> str:
     import time
 
     from google import genai
-    from google.genai.types import CreateBatchJobConfig, JobState, HttpOptions
+    from google.genai.types import CreateBatchJobConfig, HttpOptions, JobState
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     # TODO(developer): Update and un-comment below line
@@ -28,7 +28,7 @@ def generate_content(output_uri: str) -> str:
     job = client.batches.create(
         # To use a tuned model, set the model param to your tuned model using the following format:
         # model="projects/{PROJECT_ID}/locations/{LOCATION}/models/{MODEL_ID}
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         # Source link: https://storage.cloud.google.com/cloud-samples-data/batch/prompt_for_batch_gemini_predict.jsonl
         src="gs://cloud-samples-data/batch/prompt_for_batch_gemini_predict.jsonl",
         config=CreateBatchJobConfig(dest=output_uri),

genai/bounding_box/boundingbox_with_txt_img.py

@@ -102,7 +102,7 @@ def plot_bounding_boxes(image_uri: str, bounding_boxes: list[BoundingBox]) -> No
     image_uri = "https://storage.googleapis.com/generativeai-downloads/images/socks.jpg"
 
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents=[
             Part.from_uri(
                 file_uri=image_uri,

genai/content_cache/contentcache_create_with_txt_gcs_pdf.py

@@ -42,7 +42,7 @@ def create_content_cache() -> str:
     ]
 
     content_cache = client.caches.create(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         config=CreateCachedContentConfig(
             contents=contents,
             system_instruction=system_instruction,

genai/content_cache/contentcache_use_with_txt.py

@@ -22,7 +22,7 @@ def generate_content(cache_name: str) -> str:
     # Use content cache to generate text response
     # E.g cache_name = 'projects/.../locations/.../cachedContents/1111111111111111111'
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="Summarize the pdfs",
         config=GenerateContentConfig(
             cached_content=cache_name,

genai/controlled_generation/ctrlgen_with_class_schema.py

@@ -17,7 +17,6 @@ def generate_content() -> str:
     # [START googlegenaisdk_ctrlgen_with_class_schema]
     from google import genai
     from google.genai.types import GenerateContentConfig, HttpOptions
-
     from pydantic import BaseModel
 
     class Recipe(BaseModel):
@@ -26,7 +25,7 @@ class Recipe(BaseModel):
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="List a few popular cookie recipes.",
         config=GenerateContentConfig(
             response_mime_type="application/json",

genai/controlled_generation/ctrlgen_with_enum_class_schema.py

@@ -29,7 +29,7 @@ class InstrumentClass(enum.Enum):
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="What type of instrument is a guitar?",
         config={
             "response_mime_type": "text/x.enum",

genai/controlled_generation/ctrlgen_with_enum_schema.py

@@ -20,7 +20,7 @@ def generate_content() -> str:
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="What type of instrument is an oboe?",
         config=GenerateContentConfig(
             response_mime_type="text/x.enum",

genai/controlled_generation/ctrlgen_with_nested_class_schema.py

@@ -19,7 +19,6 @@ def generate_content() -> str:
 
     from google import genai
     from google.genai.types import GenerateContentConfig, HttpOptions
-
     from pydantic import BaseModel
 
     class Grade(enum.Enum):
@@ -36,7 +35,7 @@ class Recipe(BaseModel):
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="List about 10 home-baked cookies and give them grades based on tastiness.",
         config=GenerateContentConfig(
             response_mime_type="application/json",

genai/controlled_generation/ctrlgen_with_nullable_schema.py

@@ -51,7 +51,7 @@ def generate_content() -> str:
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents=prompt,
         config=GenerateContentConfig(
             response_mime_type="application/json",

genai/controlled_generation/ctrlgen_with_resp_schema.py

@@ -36,7 +36,7 @@ def generate_content() -> str:
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents=prompt,
         config={
             "response_mime_type": "application/json",

genai/count_tokens/counttoken_compute_with_txt.py

@@ -20,7 +20,7 @@ def compute_tokens_example() -> int:
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.compute_tokens(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="What's the longest word in the English language?",
     )
 

genai/count_tokens/counttoken_localtokenizer_compute_with_txt.py

@@ -17,7 +17,7 @@ def counttoken_localtokenizer_compute_with_txt() -> int:
     # [START googlegenaisdk_counttoken_localtokenizer_compute_with_txt]
     from google.genai.local_tokenizer import LocalTokenizer
 
-    tokenizer = LocalTokenizer(model_name="gemini-2.5-flash")
+    tokenizer = LocalTokenizer(model_name="gemini-3.5-flash")
     response = tokenizer.compute_tokens("What's the longest word in the English language?")
     print(response)
     # Example output:

genai/count_tokens/counttoken_localtokenizer_with_txt.py

@@ -17,7 +17,7 @@ def counttoken_localtokenizer_with_txt() -> int:
     # [START googlegenaisdk_counttoken_localtokenizer_with_txt]
     from google.genai.local_tokenizer import LocalTokenizer
 
-    tokenizer = LocalTokenizer(model_name="gemini-2.5-flash")
+    tokenizer = LocalTokenizer(model_name="gemini-3.5-flash")
     response = tokenizer.count_tokens("What's the highest mountain in Africa?")
     print(response)
     # Example output:

genai/count_tokens/counttoken_resp_with_txt.py

@@ -24,7 +24,7 @@ def count_tokens_example() -> int:
 
     # Send text to Gemini
     response = client.models.generate_content(
-        model="gemini-2.5-flash", contents=prompt
+        model="gemini-3.5-flash", contents=prompt
     )
 
     # Prompt and response tokens count

genai/count_tokens/counttoken_with_txt.py

@@ -20,7 +20,7 @@ def count_tokens() -> int:
 
     client = genai.Client(http_options=HttpOptions(api_version="v1"))
     response = client.models.count_tokens(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="What's the highest mountain in Africa?",
     )
     print(response)

genai/count_tokens/counttoken_with_txt_vid.py

@@ -29,7 +29,7 @@ def count_tokens() -> int:
     ]
 
     response = client.models.count_tokens(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents=contents,
     )
     print(response)

genai/express_mode/api_key_example.py

@@ -23,7 +23,7 @@ def generate_content() -> str:
     client = genai.Client(vertexai=True, api_key=API_KEY)
 
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="Explain bubble sort to me.",
     )
 

genai/express_mode/test_express_mode_examples.py

@@ -40,7 +40,7 @@ def test_api_key_example(mock_genai_client: MagicMock) -> None:
 
     mock_genai_client.assert_called_once_with(vertexai=True, api_key="YOUR_API_KEY")
     mock_genai_client.return_value.models.generate_content.assert_called_once_with(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="Explain bubble sort to me.",
     )
     assert response == "This is a mocked bubble sort explanation."

genai/live/live_code_exec_with_txt.py

@@ -18,11 +18,17 @@
 async def generate_content() -> list[str]:
     # [START googlegenaisdk_live_code_exec_with_txt]
     from google import genai
-    from google.genai.types import (Content, LiveConnectConfig, Modality, Part,
-                                    Tool, ToolCodeExecution)
+    from google.genai.types import (
+        Content,
+        LiveConnectConfig,
+        Modality,
+        Part,
+        Tool,
+        ToolCodeExecution,
+    )
 
     client = genai.Client()
-    model_id = "gemini-2.0-flash-live-preview-04-09"
+    model_id = "gemini-live-2.5-flash-native-audio"
     config = LiveConnectConfig(
         response_modalities=[Modality.TEXT],
         tools=[Tool(code_execution=ToolCodeExecution())],

genai/live/live_func_call_with_txt.py

@@ -20,12 +20,18 @@
 async def generate_content() -> list[FunctionResponse]:
     # [START googlegenaisdk_live_func_call_with_txt]
     from google import genai
-    from google.genai.types import (Content, FunctionDeclaration,
-                                    FunctionResponse, LiveConnectConfig,
-                                    Modality, Part, Tool)
+    from google.genai.types import (
+        Content,
+        FunctionDeclaration,
+        FunctionResponse,
+        LiveConnectConfig,
+        Modality,
+        Part,
+        Tool,
+    )
 
     client = genai.Client()
-    model_id = "gemini-2.0-flash-live-preview-04-09"
+    model_id = "gemini-live-2.5-flash-native-audio"
 
     # Simple function definitions
     turn_on_the_lights = FunctionDeclaration(name="turn_on_the_lights")

genai/live/live_ground_googsearch_with_txt.py

@@ -19,11 +19,17 @@
 async def generate_content() -> list[str]:
     # [START googlegenaisdk_live_ground_googsearch_with_txt]
     from google import genai
-    from google.genai.types import (Content, GoogleSearch, LiveConnectConfig,
-                                    Modality, Part, Tool)
+    from google.genai.types import (
+        Content,
+        GoogleSearch,
+        LiveConnectConfig,
+        Modality,
+        Part,
+        Tool,
+    )
 
     client = genai.Client()
-    model_id = "gemini-2.0-flash-live-preview-04-09"
+    model_id = "gemini-live-2.5-flash-native-audio"
     config = LiveConnectConfig(
         response_modalities=[Modality.TEXT],
         tools=[Tool(google_search=GoogleSearch())],

genai/live/live_ground_ragengine_with_txt.py

@@ -17,12 +17,19 @@
 async def generate_content(memory_corpus: str) -> list[str]:
     # [START googlegenaisdk_live_ground_ragengine_with_txt]
     from google import genai
-    from google.genai.types import (Content, LiveConnectConfig, Modality, Part,
-                                    Retrieval, Tool, VertexRagStore,
-                                    VertexRagStoreRagResource)
+    from google.genai.types import (
+        Content,
+        LiveConnectConfig,
+        Modality,
+        Part,
+        Retrieval,
+        Tool,
+        VertexRagStore,
+        VertexRagStoreRagResource,
+    )
 
     client = genai.Client()
-    model_id = "gemini-2.0-flash-live-preview-04-09"
+    model_id = "gemini-live-2.5-flash-native-audio"
     rag_store = VertexRagStore(
         rag_resources=[
             VertexRagStoreRagResource(

genai/live/live_structured_output_with_txt.py

@@ -30,8 +30,10 @@ def generate_content() -> CalendarEvent:
     import google.auth.transport.requests
     import openai
     from google.auth import default
-    from openai.types.chat import (ChatCompletionSystemMessageParam,
-                                   ChatCompletionUserMessageParam)
+    from openai.types.chat import (
+        ChatCompletionSystemMessageParam,
+        ChatCompletionUserMessageParam,
+    )
 
     project_id = os.environ["GOOGLE_CLOUD_PROJECT"]
     location = "us-central1"
@@ -59,7 +61,7 @@ def generate_content() -> CalendarEvent:
     )
 
     completion = client.beta.chat.completions.parse(
-        model="google/gemini-2.5-flash",
+        model="google/gemini-3.5-flash",
         messages=[
             ChatCompletionSystemMessageParam(
                 role="system", content="Extract the event information."

genai/live/live_txt_with_audio.py

@@ -30,7 +30,7 @@ async def generate_content() -> list[str]:
     from google.genai.types import Blob, LiveConnectConfig, Modality
 
     client = genai.Client()
-    model = "gemini-2.0-flash-live-preview-04-09"
+    model = "gemini-live-2.5-flash-native-audio"
     config = LiveConnectConfig(response_modalities=[Modality.TEXT])
 
     async with client.aio.live.connect(model=model, config=config) as session:

genai/live/live_txtgen_with_audio.py

@@ -28,7 +28,7 @@ async def generate_content() -> list[str]:
     from google.genai.types import Blob, LiveConnectConfig, Modality
 
     client = genai.Client()
-    model = "gemini-2.0-flash-live-preview-04-09"
+    model = "gemini-live-2.5-flash-native-audio"
     config = LiveConnectConfig(response_modalities=[Modality.TEXT])
 
     def get_audio(url: str) -> bytes:

genai/live/live_websocket_textgen_with_audio.py

@@ -57,7 +57,7 @@ def read_wavefile(filepath: str) -> tuple[str, str]:
     # Configuration Constants
     PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
     LOCATION = "us-central1"
-    GEMINI_MODEL_NAME = "gemini-2.0-flash-live-preview-04-09"
+    GEMINI_MODEL_NAME = "gemini-live-2.5-flash-native-audio"
     # To generate a bearer token in CLI, use:
     #   $ gcloud auth application-default print-access-token
     # It's recommended to fetch this token dynamically rather than hardcoding.

genai/live/live_websocket_textgen_with_txt.py

@@ -44,7 +44,7 @@ async def generate_content() -> str:
     # Configuration Constants
     PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
     LOCATION = "us-central1"
-    GEMINI_MODEL_NAME = "gemini-2.0-flash-live-preview-04-09"
+    GEMINI_MODEL_NAME = "gemini-live-2.5-flash-native-audio"
     # To generate a bearer token in CLI, use:
     #   $ gcloud auth application-default print-access-token
     # It's recommended to fetch this token dynamically rather than hardcoding.

genai/live/live_with_txt.py

@@ -18,11 +18,16 @@
 async def generate_content() -> list[str]:
     # [START googlegenaisdk_live_with_txt]
     from google import genai
-    from google.genai.types import (Content, HttpOptions, LiveConnectConfig,
-                                    Modality, Part)
+    from google.genai.types import (
+        Content,
+        HttpOptions,
+        LiveConnectConfig,
+        Modality,
+        Part,
+    )
 
     client = genai.Client(http_options=HttpOptions(api_version="v1beta1"))
-    model_id = "gemini-2.0-flash-live-preview-04-09"
+    model_id = "gemini-live-2.5-flash-native-audio"
 
     async with client.aio.live.connect(
         model=model_id,

genai/provisioned_throughput/provisionedthroughput_with_txt.py

@@ -31,7 +31,7 @@ def generate_content() -> str:
         )
     )
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents="How does AI work?",
     )
     print(response.text)

genai/safety/safety_with_txt.py

@@ -20,8 +20,8 @@ def generate_content() -> GenerateContentResponse:
     from google import genai
     from google.genai.types import (
         GenerateContentConfig,
-        HarmCategory,
         HarmBlockThreshold,
+        HarmCategory,
         HttpOptions,
         SafetySetting,
     )
@@ -54,7 +54,7 @@ def generate_content() -> GenerateContentResponse:
     ]
 
     response = client.models.generate_content(
-        model="gemini-2.5-flash",
+        model="gemini-3.5-flash",
         contents=prompt,
         config=GenerateContentConfig(
             system_instruction=system_instruction,