Merge branch 'main' into fix563

docs/api/google/generativeai/ChatSession.md

@@ -29,7 +29,7 @@ Contains an ongoing conversation with the model.
 <!-- Placeholder for "Used in" -->
 
 ```
->>> model = genai.GenerativeModel('models/gemini-pro')
+>>> model = genai.GenerativeModel('models/gemini-1.5-flash')
 >>> chat = model.start_chat()
 >>> response = chat.send_message("Hello")
 >>> print(response.text)
@@ -142,7 +142,7 @@ Sends the conversation history with the added message and returns the model's re
 Appends the request and response to the conversation history.
 
 ```
->>> model = genai.GenerativeModel('models/gemini-pro')
+>>> model = genai.GenerativeModel('models/gemini-1.5-flash')
 >>> chat = model.start_chat()
 >>> response = chat.send_message("Hello")
 >>> print(response.text)

docs/api/google/generativeai/GenerativeModel.md

@@ -38,7 +38,7 @@ requests. What media-types are supported for input and output is model-dependant
 >>> import google.generativeai as genai
 >>> import PIL.Image
 >>> genai.configure(api_key='YOUR_API_KEY')
->>> model = genai.GenerativeModel('models/gemini-pro')
+>>> model = genai.GenerativeModel('models/gemini-1.5-flash')
 >>> result = model.generate_content('Tell me a story about a magic backpack')
 >>> result.text
 "In the quaint little town of Lakeside, there lived a young girl named Lily..."
@@ -49,7 +49,7 @@ requests. What media-types are supported for input and output is model-dependant
 
 
 ```
->>> model = genai.GenerativeModel('models/gemini-pro')
+>>> model = genai.GenerativeModel('models/gemini-1.5-flash')
 >>> result = model.generate_content([
 ...     "Give me a recipe for these:", PIL.Image.open('scones.jpeg')])
 >>> result.text
@@ -291,7 +291,7 @@ This <a href="../../google/generativeai/GenerativeModel.md#generate_content"><co
 conversations.
 
 ```
->>> model = genai.GenerativeModel('models/gemini-pro')
+>>> model = genai.GenerativeModel('models/gemini-1.5-flash')
 >>> response = model.generate_content('Tell me a story about a magic backpack')
 >>> response.text
 ```

docs/api/google/generativeai/get_model.md

@@ -32,7 +32,7 @@ Calls the API to fetch a model by name.
 
 ```
 import pprint
-model = genai.get_model('models/gemini-pro')
+model = genai.get_model('models/gemini-1.5-flash')
 pprint.pprint(model)
 ```
 

google/generativeai/generative_models.py

@@ -36,14 +36,14 @@ class GenerativeModel:
     >>> import google.generativeai as genai
     >>> import PIL.Image
     >>> genai.configure(api_key='YOUR_API_KEY')
-    >>> model = genai.GenerativeModel('models/gemini-pro')
+    >>> model = genai.GenerativeModel('models/gemini-1.5-flash')
     >>> result = model.generate_content('Tell me a story about a magic backpack')
     >>> result.text
     "In the quaint little town of Lakeside, there lived a young girl named Lily..."
 
     Multimodal input:
 
-    >>> model = genai.GenerativeModel('models/gemini-pro')
+    >>> model = genai.GenerativeModel('models/gemini-1.5-flash')
     >>> result = model.generate_content([
     ...     "Give me a recipe for these:", PIL.Image.open('scones.jpeg')])
     >>> result.text
@@ -250,7 +250,7 @@ def generate_content(
         This `GenerativeModel.generate_content` method can handle multimodal input, and multi-turn
         conversations.
 
-        >>> model = genai.GenerativeModel('models/gemini-pro')
+        >>> model = genai.GenerativeModel('models/gemini-1.5-flash')
         >>> response = model.generate_content('Tell me a story about a magic backpack')
         >>> response.text
 
@@ -481,7 +481,7 @@ def start_chat(
 class ChatSession:
     """Contains an ongoing conversation with the model.
 
-    >>> model = genai.GenerativeModel('models/gemini-pro')
+    >>> model = genai.GenerativeModel('models/gemini-1.5-flash')
     >>> chat = model.start_chat()
     >>> response = chat.send_message("Hello")
     >>> print(response.text)
@@ -524,7 +524,7 @@ def send_message(
 
         Appends the request and response to the conversation history.
 
-        >>> model = genai.GenerativeModel('models/gemini-pro')
+        >>> model = genai.GenerativeModel('models/gemini-1.5-flash')
         >>> chat = model.start_chat()
         >>> response = chat.send_message("Hello")
         >>> print(response.text)

google/generativeai/models.py

@@ -40,7 +40,7 @@ def get_model(
 
     ```
     import pprint
-    model = genai.get_model('models/gemini-pro')
+    model = genai.get_model('models/gemini-1.5-flash')
     pprint.pprint(model)
     ```
 
@@ -112,7 +112,7 @@ def get_tuned_model(
 
     ```
     import pprint
-    model = genai.get_tuned_model('tunedModels/gemini-1.0-pro-001')
+    model = genai.get_tuned_model('tunedModels/gemini-1.5-flash')
     pprint.pprint(model)
     ```
 

google/generativeai/notebook/text_model.py

@@ -20,7 +20,7 @@
 from google.generativeai.types import generation_types
 from google.generativeai.notebook.lib import model as model_lib
 
-_DEFAULT_MODEL = "models/gemini-pro"
+_DEFAULT_MODEL = "models/gemini-1.5-flash"
 
 
 class TextModel(model_lib.AbstractModel):

samples/rest/tuned_models.sh

@@ -7,7 +7,7 @@ curl -X POST "https://generativelanguage.googleapis.com/v1beta/tunedModels?key=$
     -d '
       {
         "display_name": "number generator model",
-        "base_model": "models/gemini-1.0-pro-001",
+        "base_model": "models/gemini-1.5-flash-001-tuning",
         "tuning_task": {
           "hyperparameters": {
             "batch_size": 2,

tests/test_generative_models.py

@@ -115,7 +115,7 @@ def setUp(self):
 
     def test_hello(self):
         # Generate text from text prompt
-        model = generative_models.GenerativeModel(model_name="gemini-pro")
+        model = generative_models.GenerativeModel(model_name="gemini-1.5-flash")
 
         self.responses["generate_content"].append(simple_response("world!"))
 
@@ -138,7 +138,7 @@ def test_hello(self):
     )
     def test_image(self, content):
         # Generate text from image
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
 
         cat = "It's a cat"
         self.responses["generate_content"].append(simple_response(cat))
@@ -172,7 +172,7 @@ def test_image(self, content):
     )
     def test_generation_config_overwrite(self, config1, config2):
         # Generation config
-        model = generative_models.GenerativeModel("gemini-pro", generation_config=config1)
+        model = generative_models.GenerativeModel("gemini-1.5-flash", generation_config=config1)
 
         self.responses["generate_content"] = [
             simple_response(" world!"),
@@ -218,7 +218,7 @@ def test_generation_config_overwrite(self, config1, config2):
     )
     def test_safety_overwrite(self, safe1, safe2):
         # Safety
-        model = generative_models.GenerativeModel("gemini-pro", safety_settings=safe1)
+        model = generative_models.GenerativeModel("gemini-1.5-flash", safety_settings=safe1)
 
         self.responses["generate_content"] = [
             simple_response(" world!"),
@@ -253,7 +253,7 @@ def test_stream_basic(self):
         chunks = ["first", " second", " third"]
         self.responses["stream_generate_content"] = [(simple_response(text) for text in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Hello", stream=True)
 
         self.assertEqual(self.observed_requests[0].contents[0].parts[0].text, "Hello")
@@ -267,7 +267,7 @@ def test_stream_lookahead(self):
         chunks = ["first", " second", " third"]
         self.responses["stream_generate_content"] = [(simple_response(text) for text in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Hello", stream=True)
 
         self.assertEqual(self.observed_requests[0].contents[0].parts[0].text, "Hello")
@@ -287,7 +287,7 @@ def test_stream_prompt_feedback_blocked(self):
         ]
         self.responses["stream_generate_content"] = [(chunk for chunk in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Bad stuff!", stream=True)
 
         self.assertEqual(
@@ -322,7 +322,7 @@ def test_stream_prompt_feedback_not_blocked(self):
         ]
         self.responses["stream_generate_content"] = [(chunk for chunk in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Hello", stream=True)
 
         self.assertEqual(
@@ -389,7 +389,7 @@ def add(a: int, b: int) -> int:
 
     def test_chat(self):
         # Multi turn chat
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
 
         self.responses["generate_content"] = [
@@ -423,7 +423,7 @@ def test_chat(self):
     def test_chat_roles(self):
         self.responses["generate_content"] = [simple_response("hello!")]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
         response = chat.send_message("hello?")
         history = chat.history
@@ -792,7 +792,7 @@ def test_tool_config(self, tool_config, expected_tool_config):
         )
         self.responses["generate_content"] = [simple_response("echo echo")]
 
-        model = generative_models.GenerativeModel("gemini-pro", tools=tools)
+        model = generative_models.GenerativeModel("gemini-1.5-flash", tools=tools)
         _ = model.generate_content("Hello", tools=[tools], tool_config=tool_config)
 
         req = self.observed_requests[0]
@@ -811,7 +811,9 @@ def test_tool_config(self, tool_config, expected_tool_config):
     )
     def test_system_instruction(self, instruction, expected_instr):
         self.responses["generate_content"] = [simple_response("echo echo")]
-        model = generative_models.GenerativeModel("gemini-pro", system_instruction=instruction)
+        model = generative_models.GenerativeModel(
+            "gemini-1.5-flash", system_instruction=instruction
+        )
 
         _ = model.generate_content("test")
 
@@ -852,7 +854,7 @@ def test_count_tokens_smoke(self, kwargs):
         )
 
     def test_repr_for_unary_non_streamed_response(self):
-        model = generative_models.GenerativeModel(model_name="gemini-pro")
+        model = generative_models.GenerativeModel(model_name="gemini-1.5-flash")
         self.responses["generate_content"].append(simple_response("world!"))
         response = model.generate_content("Hello")
 
@@ -885,7 +887,7 @@ def test_repr_for_streaming_start_to_finish(self):
         chunks = ["first", " second", " third"]
         self.responses["stream_generate_content"] = [(simple_response(text) for text in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Hello", stream=True)
         iterator = iter(response)
 
@@ -980,7 +982,7 @@ def test_repr_error_info_for_stream_prompt_feedback_blocked(self):
         ]
         self.responses["stream_generate_content"] = [(chunk for chunk in chunks)]
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         response = model.generate_content("Bad stuff!", stream=True)
 
         result = repr(response)
@@ -1096,7 +1098,7 @@ def test_repr_error_info_for_chat_streaming_unexpected_stop(self):
 
     def test_repr_for_multi_turn_chat(self):
         # Multi turn chat
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
 
         self.responses["generate_content"] = [
@@ -1119,7 +1121,7 @@ def test_repr_for_multi_turn_chat(self):
             """\
             ChatSession(
                 model=genai.GenerativeModel(
-                    model_name='models/gemini-pro',
+                    model_name='models/gemini-1.5-flash',
                     generation_config={},
                     safety_settings={},
                     tools=None,
@@ -1133,7 +1135,7 @@ def test_repr_for_multi_turn_chat(self):
 
     def test_repr_for_incomplete_streaming_chat(self):
         # Multi turn chat
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
 
         self.responses["stream_generate_content"] = [
@@ -1148,7 +1150,7 @@ def test_repr_for_incomplete_streaming_chat(self):
             """\
             ChatSession(
                 model=genai.GenerativeModel(
-                    model_name='models/gemini-pro',
+                    model_name='models/gemini-1.5-flash',
                     generation_config={},
                     safety_settings={},
                     tools=None,
@@ -1162,7 +1164,7 @@ def test_repr_for_incomplete_streaming_chat(self):
 
     def test_repr_for_broken_streaming_chat(self):
         # Multi turn chat
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
 
         self.responses["stream_generate_content"] = [
@@ -1193,7 +1195,7 @@ def test_repr_for_broken_streaming_chat(self):
             """\
             ChatSession(
                 model=genai.GenerativeModel(
-                    model_name='models/gemini-pro',
+                    model_name='models/gemini-1.5-flash',
                     generation_config={},
                     safety_settings={},
                     tools=None,
@@ -1206,7 +1208,9 @@ def test_repr_for_broken_streaming_chat(self):
         self.assertEqual(expected, result)
 
     def test_repr_for_system_instruction(self):
-        model = generative_models.GenerativeModel("gemini-pro", system_instruction="Be excellent.")
+        model = generative_models.GenerativeModel(
+            "gemini-1.5-flash", system_instruction="Be excellent."
+        )
         result = repr(model)
         self.assertIn("system_instruction='Be excellent.'", result)
 
@@ -1237,7 +1241,7 @@ def test_chat_with_request_options(self):
         )
         request_options = {"timeout": 120}
 
-        model = generative_models.GenerativeModel("gemini-pro")
+        model = generative_models.GenerativeModel("gemini-1.5-flash")
         chat = model.start_chat()
         chat.send_message("hello", request_options=helper_types.RequestOptions(**request_options))
 

tests/test_generative_models_async.py

@@ -80,7 +80,7 @@ async def count_tokens(
 
     async def test_basic(self):
         # Generate text from text prompt
-        model = generative_models.GenerativeModel(model_name="gemini-pro")
+        model = generative_models.GenerativeModel(model_name="gemini-1.5-flash")
 
         self.responses["generate_content"] = [simple_response("world!")]
 
@@ -93,7 +93,7 @@ async def test_basic(self):
 
     async def test_streaming(self):
         # Generate text from text prompt
-        model = generative_models.GenerativeModel(model_name="gemini-pro")
+        model = generative_models.GenerativeModel(model_name="gemini-1.5-flash")
 
         async def responses():
             for c in "world!":
@@ -195,7 +195,7 @@ async def test_tool_config(self, tool_config, expected_tool_config):
         )
         self.responses["generate_content"] = [simple_response("echo echo")]
 
-        model = generative_models.GenerativeModel("gemini-pro", tools=tools)
+        model = generative_models.GenerativeModel("gemini-1.5-flash", tools=tools)
         _ = await model.generate_content_async("Hello", tools=[tools], tool_config=tool_config)
 
         req = self.observed_requests[0]