google-genai 0.0.1__py3-none-any.whl → 0.2.0__py3-none-any.whl

google/genai/live.py

@@ -53,6 +53,12 @@ except ModuleNotFoundError:
   from websockets.client import connect
 
 
+_FUNCTION_RESPONSE_REQUIRES_ID = (
+    'FunctionResponse request must have an `id` field from the'
+    ' response of a ToolCall.FunctionalCalls in Google AI.'
+)
+
+
 class AsyncSession:
   """AsyncSession."""
 
@@ -81,20 +87,23 @@ class AsyncSession:
     """Receive model responses from the server.
 
     The method will yield the model responses from the server. The returned
-    responses will represent a complete model turn.
-    when the returned message is fuction call, user must call `send` with the
-    function response to continue the turn.
-    Example usage:
-    ```
-    client = genai.Client(api_key=API_KEY)
-
-    async with client.aio.live.connect(model='...') as session:
-      await session.send(input='Hello world!', end_of_turn=True)
-      async for message in session.receive():
-        print(message)
-    ```
+    responses will represent a complete model turn. When the returned message
+    is fuction call, user must call `send` with the function response to
+    continue the turn.
+
     Yields:
-        The model responses from the server.
+      The model responses from the server.
+
+    Example usage:
+
+    .. code-block:: python
+
+      client = genai.Client(api_key=API_KEY)
+
+      async with client.aio.live.connect(model='...') as session:
+        await session.send(input='Hello world!', end_of_turn=True)
+        async for message in session.receive():
+          print(message)
     """
     # TODO(b/365983264) Handle intermittent issues for the user.
     while result := await self._receive():
@@ -113,28 +122,27 @@ class AsyncSession:
     input stream to the model and the other task will be used to receive the
     responses from the model.
 
-    Example usage:
-    ```
-    client = genai.Client(api_key=API_KEY)
-    config = {'response_modalities': ['AUDIO']}
-
-    async def audio_stream():
-      stream = read_audio()
-      for data in stream:
-        yield data
-
-    async with client.aio.live.connect(model='...') as session:
-      for audio in session.start_stream(stream = audio_stream(),
-      mime_type = 'audio/pcm'):
-        play_audio_chunk(audio.data)
-    ```
-
     Args:
-        stream: An iterator that yields the model response.
-        mime_type: The MIME type of the data in the stream.
+      stream: An iterator that yields the model response.
+      mime_type: The MIME type of the data in the stream.
 
     Yields:
-        The audio bytes received from the model and server response messages.
+      The audio bytes received from the model and server response messages.
+
+    Example usage:
+
+    .. code-block:: python
+
+      client = genai.Client(api_key=API_KEY)
+      config = {'response_modalities': ['AUDIO']}
+      async def audio_stream():
+        stream = read_audio()
+        for data in stream:
+          yield data
+      async with client.aio.live.connect(model='...') as session:
+        for audio in session.start_stream(stream = audio_stream(),
+        mime_type = 'audio/pcm'):
+          play_audio_chunk(audio.data)
     """
     stop_event = asyncio.Event()
     # Start the send loop. When stream is complete stop_event is set.
@@ -340,6 +348,8 @@ class AsyncSession:
       input = [input]
     elif isinstance(input, dict) and 'name' in input and 'response' in input:
       # ToolResponse.FunctionResponse
+      if not (self._api_client.vertexai) and 'id' not in input:
+        raise ValueError(_FUNCTION_RESPONSE_REQUIRES_ID)
       input = [input]
 
     if isinstance(input, Sequence) and any(
@@ -395,7 +405,29 @@ class AsyncSession:
       client_message = {'client_content': input.model_dump(exclude_none=True)}
     elif isinstance(input, types.LiveClientToolResponse):
       # ToolResponse.FunctionResponse
+      if not (self._api_client.vertexai) and not (input.function_responses[0].id):
+        raise ValueError(_FUNCTION_RESPONSE_REQUIRES_ID)
       client_message = {'tool_response': input.model_dump(exclude_none=True)}
+    elif isinstance(input, types.FunctionResponse):
+      if not (self._api_client.vertexai) and not (input.id):
+        raise ValueError(_FUNCTION_RESPONSE_REQUIRES_ID)
+      client_message = {
+          'tool_response': {
+              'function_responses': [input.model_dump(exclude_none=True)]
+          }
+      }
+    elif isinstance(input, Sequence) and isinstance(
+        input[0], types.FunctionResponse
+    ):
+      if not (self._api_client.vertexai) and not (input[0].id):
+        raise ValueError(_FUNCTION_RESPONSE_REQUIRES_ID)
+      client_message = {
+          'tool_response': {
+              'function_responses': [
+                  c.model_dump(exclude_none=True) for c in input
+              ]
+          }
+      }
     else:
       raise ValueError(
           f'Unsupported input type "{type(input)}" or input content "{input}"'
@@ -571,16 +603,16 @@ class AsyncLive(_common.BaseModule):
   ) -> AsyncSession:
     """Connect to the live server.
 
-    Example usage:
-    ```
-    client = genai.Client(api_key=API_KEY)
-    config = {}
-
-    async with client.aio.live.connect(model='gemini-1.0-pro-002', config=config) as session:
-      await session.send(input='Hello world!', end_of_turn=True)
-      async for message in session:
-        print(message)
-    ```
+    Usage:
+
+    .. code-block:: python
+
+      client = genai.Client(api_key=API_KEY)
+      config = {}
+      async with client.aio.live.connect(model='gemini-1.0-pro-002', config=config) as session:
+        await session.send(input='Hello world!', end_of_turn=True)
+        async for message in session:
+          print(message)
     """
     base_url = self.api_client._websocket_base_url()
     if self.api_client.api_key:

google/genai/models.py

@@ -880,6 +880,11 @@ def _GenerateContentConfig_to_mldev(
         getv(from_object, ['response_modalities']),
     )
 
+  if getv(from_object, ['media_resolution']):
+    raise ValueError(
+        'media_resolution parameter is not supported in Google AI.'
+    )
+
   if getv(from_object, ['speech_config']) is not None:
     setv(
         to_object,
@@ -1022,6 +1027,11 @@ def _GenerateContentConfig_to_vertex(
         getv(from_object, ['response_modalities']),
     )
 
+  if getv(from_object, ['media_resolution']) is not None:
+    setv(
+        to_object, ['mediaResolution'], getv(from_object, ['media_resolution'])
+    )
+
   if getv(from_object, ['speech_config']) is not None:
     setv(
         to_object,
@@ -1116,8 +1126,15 @@ def _EmbedContentConfig_to_mldev(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['task_type']) is not None:
-    setv(parent_object, ['requests[]'], getv(from_object, ['task_type']))
+    setv(
+        parent_object,
+        ['requests[]', 'taskType'],
+        getv(from_object, ['task_type']),
+    )
 
   if getv(from_object, ['title']) is not None:
     setv(parent_object, ['requests[]', 'title'], getv(from_object, ['title']))
@@ -1144,10 +1161,13 @@ def _EmbedContentConfig_to_vertex(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['task_type']) is not None:
     setv(
         parent_object,
-        ['instances[]', 'taskType'],
+        ['instances[]', 'task_type'],
         getv(from_object, ['task_type']),
     )
 
@@ -1253,6 +1273,9 @@ def _GenerateImageConfig_to_mldev(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['output_gcs_uri']):
     raise ValueError('output_gcs_uri parameter is not supported in Google AI.')
 
@@ -1348,6 +1371,9 @@ def _GenerateImageConfig_to_vertex(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['output_gcs_uri']) is not None:
     setv(
         parent_object,
@@ -1778,6 +1804,9 @@ def _EditImageConfig_to_mldev(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['output_gcs_uri']):
     raise ValueError('output_gcs_uri parameter is not supported in Google AI.')
 
@@ -1870,6 +1899,9 @@ def _EditImageConfig_to_vertex(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['output_gcs_uri']) is not None:
     setv(
         parent_object,
@@ -2042,6 +2074,9 @@ def _UpscaleImageAPIConfig_to_mldev(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['upscale_factor']) is not None:
     setv(
         parent_object,
@@ -2089,6 +2124,9 @@ def _UpscaleImageAPIConfig_to_vertex(
     parent_object: dict = None,
 ) -> dict:
   to_object = {}
+  if getv(from_object, ['http_options']) is not None:
+    setv(to_object, ['httpOptions'], getv(from_object, ['http_options']))
+
   if getv(from_object, ['upscale_factor']) is not None:
     setv(
         parent_object,
@@ -3695,6 +3733,29 @@ class Models(_common.BaseModule):
       contents: Union[types.ContentListUnion, types.ContentListUnionDict],
       config: Optional[types.EmbedContentConfigOrDict] = None,
   ) -> types.EmbedContentResponse:
+    """Calculates embeddings for the given contents(only text is supported).
+
+    Args:
+      model (str): The model to use.
+      contents (list[Content]): The contents to embed.
+      config (EmbedContentConfig): Optional configuration for embeddings.
+
+    Usage:
+
+    .. code-block:: python
+
+      embeddings = client.models.embed_content(
+          model= 'text-embedding-004',
+          contents=[
+              'What is your name?',
+              'What is your favorite color?',
+          ],
+          config={
+              'output_dimensionality': 64
+          },
+      )
+    """
+
     parameter_model = types._EmbedContentParameters(
         model=model,
         contents=contents,
@@ -4327,6 +4388,7 @@ class Models(_common.BaseModule):
     logging.info(
         f'AFC is enabled with max remote calls: {remaining_remote_calls_afc}.'
     )
+    automatic_function_calling_history = []
     while remaining_remote_calls_afc > 0:
       response = self._generate_content(
           model=model, contents=contents, config=config
@@ -4351,11 +4413,17 @@ class Models(_common.BaseModule):
         break
       contents = t.t_contents(self.api_client, contents)
       contents.append(response.candidates[0].content)
-      contents.append({
-          'role': 'user',
-          'parts': func_response_parts,
-      })
-
+      contents.append(
+          types.Content(
+              role='user',
+              parts=func_response_parts,
+          )
+      )
+      automatic_function_calling_history.extend(contents)
+    if _extra_utils.should_append_afc_history(config):
+      response.automatic_function_calling_history = (
+          automatic_function_calling_history
+      )
     return response
 
   def upscale_image(
@@ -4563,6 +4631,29 @@ class AsyncModels(_common.BaseModule):
       contents: Union[types.ContentListUnion, types.ContentListUnionDict],
       config: Optional[types.EmbedContentConfigOrDict] = None,
   ) -> types.EmbedContentResponse:
+    """Calculates embeddings for the given contents(only text is supported).
+
+    Args:
+      model (str): The model to use.
+      contents (list[Content]): The contents to embed.
+      config (EmbedContentConfig): Optional configuration for embeddings.
+
+    Usage:
+
+    .. code-block:: python
+
+      embeddings = client.models.embed_content(
+          model= 'text-embedding-004',
+          contents=[
+              'What is your name?',
+              'What is your favorite color?',
+          ],
+          config={
+              'output_dimensionality': 64
+          },
+      )
+    """
+
     parameter_model = types._EmbedContentParameters(
         model=model,
         contents=contents,
@@ -5185,6 +5276,7 @@ class AsyncModels(_common.BaseModule):
     logging.info(
         f'AFC is enabled with max remote calls: {remaining_remote_calls_afc}.'
     )
+    automatic_function_calling_history = []
     while remaining_remote_calls_afc > 0:
       response = await self._generate_content(
           model=model, contents=contents, config=config
@@ -5209,11 +5301,18 @@ class AsyncModels(_common.BaseModule):
         break
       contents = t.t_contents(self.api_client, contents)
       contents.append(response.candidates[0].content)
-      contents.append({
-          'role': 'user',
-          'parts': func_response_parts,
-      })
+      contents.append(
+          types.Content(
+              role='user',
+              parts=func_response_parts,
+          )
+      )
+      automatic_function_calling_history.extend(contents)
 
+    if _extra_utils.should_append_afc_history(config):
+      response.automatic_function_calling_history = (
+          automatic_function_calling_history
+      )
     return response
 
   async def list(