As of January 1, 2020 this library no longer supports Python 2 on the latest released version.
Library versions released prior to that date will continue to be available. For more information please
visit Python 2 support on Google Cloud.
Source code for google.ai.generativelanguage_v1alpha.types.generative_service
# -*- coding: utf-8 -*-
# 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
#
# http://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.
#
from __future__ import annotations
from typing import MutableMapping, MutableSequence
import proto # type: ignore
from google.ai.generativelanguage_v1alpha.types import citation, retriever, safety
from google.ai.generativelanguage_v1alpha.types import content as gag_content
__protobuf__ = proto.module(
package="google.ai.generativelanguage.v1alpha",
manifest={
"TaskType",
"GenerateContentRequest",
"PrebuiltVoiceConfig",
"VoiceConfig",
"SpeechConfig",
"GenerationConfig",
"SemanticRetrieverConfig",
"GenerateContentResponse",
"Candidate",
"LogprobsResult",
"AttributionSourceId",
"GroundingAttribution",
"RetrievalMetadata",
"GroundingMetadata",
"SearchEntryPoint",
"GroundingChunk",
"Segment",
"GroundingSupport",
"GenerateAnswerRequest",
"GenerateAnswerResponse",
"EmbedContentRequest",
"ContentEmbedding",
"EmbedContentResponse",
"BatchEmbedContentsRequest",
"BatchEmbedContentsResponse",
"CountTokensRequest",
"CountTokensResponse",
"BidiGenerateContentSetup",
"BidiGenerateContentClientContent",
"BidiGenerateContentRealtimeInput",
"BidiGenerateContentToolResponse",
"BidiGenerateContentClientMessage",
"BidiGenerateContentSetupComplete",
"BidiGenerateContentServerContent",
"BidiGenerateContentToolCall",
"BidiGenerateContentToolCallCancellation",
"BidiGenerateContentServerMessage",
},
)
[docs]class TaskType(proto.Enum):
r"""Type of task for which the embedding will be used.
Values:
TASK_TYPE_UNSPECIFIED (0):
Unset value, which will default to one of the
other enum values.
RETRIEVAL_QUERY (1):
Specifies the given text is a query in a
search/retrieval setting.
RETRIEVAL_DOCUMENT (2):
Specifies the given text is a document from
the corpus being searched.
SEMANTIC_SIMILARITY (3):
Specifies the given text will be used for
STS.
CLASSIFICATION (4):
Specifies that the given text will be
classified.
CLUSTERING (5):
Specifies that the embeddings will be used
for clustering.
QUESTION_ANSWERING (6):
Specifies that the given text will be used
for question answering.
FACT_VERIFICATION (7):
Specifies that the given text will be used
for fact verification.
"""
TASK_TYPE_UNSPECIFIED = 0
RETRIEVAL_QUERY = 1
RETRIEVAL_DOCUMENT = 2
SEMANTIC_SIMILARITY = 3
CLASSIFICATION = 4
CLUSTERING = 5
QUESTION_ANSWERING = 6
FACT_VERIFICATION = 7
[docs]class GenerateContentRequest(proto.Message):
r"""Request to generate a completion from the model.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
model (str):
Required. The name of the ``Model`` to use for generating
the completion.
Format: ``models/{model}``.
system_instruction (google.ai.generativelanguage_v1alpha.types.Content):
Optional. Developer set `system
instruction(s) <https://ai.google.dev/gemini-api/docs/system-instructions>`__.
Currently, text only.
This field is a member of `oneof`_ ``_system_instruction``.
contents (MutableSequence[google.ai.generativelanguage_v1alpha.types.Content]):
Required. The content of the current conversation with the
model.
For single-turn queries, this is a single instance. For
multi-turn queries like
`chat <https://ai.google.dev/gemini-api/docs/text-generation#chat>`__,
this is a repeated field that contains the conversation
history and the latest request.
tools (MutableSequence[google.ai.generativelanguage_v1alpha.types.Tool]):
Optional. A list of ``Tools`` the ``Model`` may use to
generate the next response.
A ``Tool`` is a piece of code that enables the system to
interact with external systems to perform an action, or set
of actions, outside of knowledge and scope of the ``Model``.
Supported ``Tool``\ s are ``Function`` and
``code_execution``. Refer to the `Function
calling <https://ai.google.dev/gemini-api/docs/function-calling>`__
and the `Code
execution <https://ai.google.dev/gemini-api/docs/code-execution>`__
guides to learn more.
tool_config (google.ai.generativelanguage_v1alpha.types.ToolConfig):
Optional. Tool configuration for any ``Tool`` specified in
the request. Refer to the `Function calling
guide <https://ai.google.dev/gemini-api/docs/function-calling#function_calling_mode>`__
for a usage example.
safety_settings (MutableSequence[google.ai.generativelanguage_v1alpha.types.SafetySetting]):
Optional. A list of unique ``SafetySetting`` instances for
blocking unsafe content.
This will be enforced on the
``GenerateContentRequest.contents`` and
``GenerateContentResponse.candidates``. There should not be
more than one setting for each ``SafetyCategory`` type. The
API will block any contents and responses that fail to meet
the thresholds set by these settings. This list overrides
the default settings for each ``SafetyCategory`` specified
in the safety_settings. If there is no ``SafetySetting`` for
a given ``SafetyCategory`` provided in the list, the API
will use the default safety setting for that category. Harm
categories HARM_CATEGORY_HATE_SPEECH,
HARM_CATEGORY_SEXUALLY_EXPLICIT,
HARM_CATEGORY_DANGEROUS_CONTENT, HARM_CATEGORY_HARASSMENT,
HARM_CATEGORY_CIVIC_INTEGRITY are supported. Refer to the
`guide <https://ai.google.dev/gemini-api/docs/safety-settings>`__
for detailed information on available safety settings. Also
refer to the `Safety
guidance <https://ai.google.dev/gemini-api/docs/safety-guidance>`__
to learn how to incorporate safety considerations in your AI
applications.
generation_config (google.ai.generativelanguage_v1alpha.types.GenerationConfig):
Optional. Configuration options for model
generation and outputs.
This field is a member of `oneof`_ ``_generation_config``.
cached_content (str):
Optional. The name of the content
`cached <https://ai.google.dev/gemini-api/docs/caching>`__
to use as context to serve the prediction. Format:
``cachedContents/{cachedContent}``
This field is a member of `oneof`_ ``_cached_content``.
"""
model: str = proto.Field(
proto.STRING,
number=1,
)
system_instruction: gag_content.Content = proto.Field(
proto.MESSAGE,
number=8,
optional=True,
message=gag_content.Content,
)
contents: MutableSequence[gag_content.Content] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
tools: MutableSequence[gag_content.Tool] = proto.RepeatedField(
proto.MESSAGE,
number=5,
message=gag_content.Tool,
)
tool_config: gag_content.ToolConfig = proto.Field(
proto.MESSAGE,
number=7,
message=gag_content.ToolConfig,
)
safety_settings: MutableSequence[safety.SafetySetting] = proto.RepeatedField(
proto.MESSAGE,
number=3,
message=safety.SafetySetting,
)
generation_config: "GenerationConfig" = proto.Field(
proto.MESSAGE,
number=4,
optional=True,
message="GenerationConfig",
)
cached_content: str = proto.Field(
proto.STRING,
number=9,
optional=True,
)
[docs]class PrebuiltVoiceConfig(proto.Message):
r"""The configuration for the prebuilt speaker to use.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
voice_name (str):
The name of the preset voice to use.
This field is a member of `oneof`_ ``_voice_name``.
"""
voice_name: str = proto.Field(
proto.STRING,
number=1,
optional=True,
)
[docs]class VoiceConfig(proto.Message):
r"""The configuration for the voice to use.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
prebuilt_voice_config (google.ai.generativelanguage_v1alpha.types.PrebuiltVoiceConfig):
The configuration for the prebuilt voice to
use.
This field is a member of `oneof`_ ``voice_config``.
"""
prebuilt_voice_config: "PrebuiltVoiceConfig" = proto.Field(
proto.MESSAGE,
number=1,
oneof="voice_config",
message="PrebuiltVoiceConfig",
)
[docs]class SpeechConfig(proto.Message):
r"""The speech generation config.
Attributes:
voice_config (google.ai.generativelanguage_v1alpha.types.VoiceConfig):
The configuration for the speaker to use.
"""
voice_config: "VoiceConfig" = proto.Field(
proto.MESSAGE,
number=1,
message="VoiceConfig",
)
[docs]class GenerationConfig(proto.Message):
r"""Configuration options for model generation and outputs. Not
all parameters are configurable for every model.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
candidate_count (int):
Optional. Number of generated responses to
return.
Currently, this value can only be set to 1. If
unset, this will default to 1.
This field is a member of `oneof`_ ``_candidate_count``.
stop_sequences (MutableSequence[str]):
Optional. The set of character sequences (up to 5) that will
stop output generation. If specified, the API will stop at
the first appearance of a ``stop_sequence``. The stop
sequence will not be included as part of the response.
max_output_tokens (int):
Optional. The maximum number of tokens to include in a
response candidate.
Note: The default value varies by model, see the
``Model.output_token_limit`` attribute of the ``Model``
returned from the ``getModel`` function.
This field is a member of `oneof`_ ``_max_output_tokens``.
temperature (float):
Optional. Controls the randomness of the output.
Note: The default value varies by model, see the
``Model.temperature`` attribute of the ``Model`` returned
from the ``getModel`` function.
Values can range from [0.0, 2.0].
This field is a member of `oneof`_ ``_temperature``.
top_p (float):
Optional. The maximum cumulative probability of tokens to
consider when sampling.
The model uses combined Top-k and Top-p (nucleus) sampling.
Tokens are sorted based on their assigned probabilities so
that only the most likely tokens are considered. Top-k
sampling directly limits the maximum number of tokens to
consider, while Nucleus sampling limits the number of tokens
based on the cumulative probability.
Note: The default value varies by ``Model`` and is specified
by the\ ``Model.top_p`` attribute returned from the
``getModel`` function. An empty ``top_k`` attribute
indicates that the model doesn't apply top-k sampling and
doesn't allow setting ``top_k`` on requests.
This field is a member of `oneof`_ ``_top_p``.
top_k (int):
Optional. The maximum number of tokens to consider when
sampling.
Gemini models use Top-p (nucleus) sampling or a combination
of Top-k and nucleus sampling. Top-k sampling considers the
set of ``top_k`` most probable tokens. Models running with
nucleus sampling don't allow top_k setting.
Note: The default value varies by ``Model`` and is specified
by the\ ``Model.top_p`` attribute returned from the
``getModel`` function. An empty ``top_k`` attribute
indicates that the model doesn't apply top-k sampling and
doesn't allow setting ``top_k`` on requests.
This field is a member of `oneof`_ ``_top_k``.
response_mime_type (str):
Optional. MIME type of the generated candidate text.
Supported MIME types are: ``text/plain``: (default) Text
output. ``application/json``: JSON response in the response
candidates. ``text/x.enum``: ENUM as a string response in
the response candidates. Refer to the
`docs <https://ai.google.dev/gemini-api/docs/prompting_with_media#plain_text_formats>`__
for a list of all supported text MIME types.
response_schema (google.ai.generativelanguage_v1alpha.types.Schema):
Optional. Output schema of the generated candidate text.
Schemas must be a subset of the `OpenAPI
schema <https://spec.openapis.org/oas/v3.0.3#schema>`__ and
can be objects, primitives or arrays.
If set, a compatible ``response_mime_type`` must also be
set. Compatible MIME types: ``application/json``: Schema for
JSON response. Refer to the `JSON text generation
guide <https://ai.google.dev/gemini-api/docs/json-mode>`__
for more details.
presence_penalty (float):
Optional. Presence penalty applied to the next token's
logprobs if the token has already been seen in the response.
This penalty is binary on/off and not dependant on the
number of times the token is used (after the first). Use
[frequency_penalty][google.ai.generativelanguage.v1alpha.GenerationConfig.frequency_penalty]
for a penalty that increases with each use.
A positive penalty will discourage the use of tokens that
have already been used in the response, increasing the
vocabulary.
A negative penalty will encourage the use of tokens that
have already been used in the response, decreasing the
vocabulary.
This field is a member of `oneof`_ ``_presence_penalty``.
frequency_penalty (float):
Optional. Frequency penalty applied to the next token's
logprobs, multiplied by the number of times each token has
been seen in the respponse so far.
A positive penalty will discourage the use of tokens that
have already been used, proportional to the number of times
the token has been used: The more a token is used, the more
dificult it is for the model to use that token again
increasing the vocabulary of responses.
Caution: A *negative* penalty will encourage the model to
reuse tokens proportional to the number of times the token
has been used. Small negative values will reduce the
vocabulary of a response. Larger negative values will cause
the model to start repeating a common token until it hits
the
[max_output_tokens][google.ai.generativelanguage.v1alpha.GenerationConfig.max_output_tokens]
limit.
This field is a member of `oneof`_ ``_frequency_penalty``.
response_logprobs (bool):
Optional. If true, export the logprobs
results in response.
This field is a member of `oneof`_ ``_response_logprobs``.
logprobs (int):
Optional. Only valid if
[response_logprobs=True][google.ai.generativelanguage.v1alpha.GenerationConfig.response_logprobs].
This sets the number of top logprobs to return at each
decoding step in the
[Candidate.logprobs_result][google.ai.generativelanguage.v1alpha.Candidate.logprobs_result].
This field is a member of `oneof`_ ``_logprobs``.
enable_enhanced_civic_answers (bool):
Optional. Enables enhanced civic answers. It
may not be available for all models.
This field is a member of `oneof`_ ``_enable_enhanced_civic_answers``.
response_modalities (MutableSequence[google.ai.generativelanguage_v1alpha.types.GenerationConfig.Modality]):
Optional. The requested modalities of the
response. Represents the set of modalities that
the model can return, and should be expected in
the response. This is an exact match to the
modalities of the response.
A model may have multiple combinations of
supported modalities. If the requested
modalities do not match any of the supported
combinations, an error will be returned.
An empty list is equivalent to requesting only
text.
speech_config (google.ai.generativelanguage_v1alpha.types.SpeechConfig):
Optional. The speech generation config.
This field is a member of `oneof`_ ``_speech_config``.
"""
[docs] class Modality(proto.Enum):
r"""Supported modalities of the response.
Values:
MODALITY_UNSPECIFIED (0):
Default value.
TEXT (1):
Indicates the model should return text.
IMAGE (2):
Indicates the model should return images.
AUDIO (3):
Indicates the model should return audio.
"""
MODALITY_UNSPECIFIED = 0
TEXT = 1
IMAGE = 2
AUDIO = 3
candidate_count: int = proto.Field(
proto.INT32,
number=1,
optional=True,
)
stop_sequences: MutableSequence[str] = proto.RepeatedField(
proto.STRING,
number=2,
)
max_output_tokens: int = proto.Field(
proto.INT32,
number=4,
optional=True,
)
temperature: float = proto.Field(
proto.FLOAT,
number=5,
optional=True,
)
top_p: float = proto.Field(
proto.FLOAT,
number=6,
optional=True,
)
top_k: int = proto.Field(
proto.INT32,
number=7,
optional=True,
)
response_mime_type: str = proto.Field(
proto.STRING,
number=13,
)
response_schema: gag_content.Schema = proto.Field(
proto.MESSAGE,
number=14,
message=gag_content.Schema,
)
presence_penalty: float = proto.Field(
proto.FLOAT,
number=15,
optional=True,
)
frequency_penalty: float = proto.Field(
proto.FLOAT,
number=16,
optional=True,
)
response_logprobs: bool = proto.Field(
proto.BOOL,
number=17,
optional=True,
)
logprobs: int = proto.Field(
proto.INT32,
number=18,
optional=True,
)
enable_enhanced_civic_answers: bool = proto.Field(
proto.BOOL,
number=19,
optional=True,
)
response_modalities: MutableSequence[Modality] = proto.RepeatedField(
proto.ENUM,
number=20,
enum=Modality,
)
speech_config: "SpeechConfig" = proto.Field(
proto.MESSAGE,
number=21,
optional=True,
message="SpeechConfig",
)
[docs]class SemanticRetrieverConfig(proto.Message):
r"""Configuration for retrieving grounding content from a ``Corpus`` or
``Document`` created using the Semantic Retriever API.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
source (str):
Required. Name of the resource for retrieval. Example:
``corpora/123`` or ``corpora/123/documents/abc``.
query (google.ai.generativelanguage_v1alpha.types.Content):
Required. Query to use for matching ``Chunk``\ s in the
given resource by similarity.
metadata_filters (MutableSequence[google.ai.generativelanguage_v1alpha.types.MetadataFilter]):
Optional. Filters for selecting ``Document``\ s and/or
``Chunk``\ s from the resource.
max_chunks_count (int):
Optional. Maximum number of relevant ``Chunk``\ s to
retrieve.
This field is a member of `oneof`_ ``_max_chunks_count``.
minimum_relevance_score (float):
Optional. Minimum relevance score for retrieved relevant
``Chunk``\ s.
This field is a member of `oneof`_ ``_minimum_relevance_score``.
"""
source: str = proto.Field(
proto.STRING,
number=1,
)
query: gag_content.Content = proto.Field(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
metadata_filters: MutableSequence[retriever.MetadataFilter] = proto.RepeatedField(
proto.MESSAGE,
number=3,
message=retriever.MetadataFilter,
)
max_chunks_count: int = proto.Field(
proto.INT32,
number=4,
optional=True,
)
minimum_relevance_score: float = proto.Field(
proto.FLOAT,
number=5,
optional=True,
)
[docs]class GenerateContentResponse(proto.Message):
r"""Response from the model supporting multiple candidate responses.
Safety ratings and content filtering are reported for both prompt in
``GenerateContentResponse.prompt_feedback`` and for each candidate
in ``finish_reason`` and in ``safety_ratings``. The API:
- Returns either all requested candidates or none of them
- Returns no candidates at all only if there was something wrong
with the prompt (check ``prompt_feedback``)
- Reports feedback on each candidate in ``finish_reason`` and
``safety_ratings``.
Attributes:
candidates (MutableSequence[google.ai.generativelanguage_v1alpha.types.Candidate]):
Candidate responses from the model.
prompt_feedback (google.ai.generativelanguage_v1alpha.types.GenerateContentResponse.PromptFeedback):
Returns the prompt's feedback related to the
content filters.
usage_metadata (google.ai.generativelanguage_v1alpha.types.GenerateContentResponse.UsageMetadata):
Output only. Metadata on the generation
requests' token usage.
model_version (str):
Output only. The model version used to
generate the response.
"""
[docs] class PromptFeedback(proto.Message):
r"""A set of the feedback metadata the prompt specified in
``GenerateContentRequest.content``.
Attributes:
block_reason (google.ai.generativelanguage_v1alpha.types.GenerateContentResponse.PromptFeedback.BlockReason):
Optional. If set, the prompt was blocked and
no candidates are returned. Rephrase the prompt.
safety_ratings (MutableSequence[google.ai.generativelanguage_v1alpha.types.SafetyRating]):
Ratings for safety of the prompt.
There is at most one rating per category.
"""
[docs] class BlockReason(proto.Enum):
r"""Specifies the reason why the prompt was blocked.
Values:
BLOCK_REASON_UNSPECIFIED (0):
Default value. This value is unused.
SAFETY (1):
Prompt was blocked due to safety reasons. Inspect
``safety_ratings`` to understand which safety category
blocked it.
OTHER (2):
Prompt was blocked due to unknown reasons.
BLOCKLIST (3):
Prompt was blocked due to the terms which are
included from the terminology blocklist.
PROHIBITED_CONTENT (4):
Prompt was blocked due to prohibited content.
IMAGE_SAFETY (5):
Candidates blocked due to unsafe image
generation content.
"""
BLOCK_REASON_UNSPECIFIED = 0
SAFETY = 1
OTHER = 2
BLOCKLIST = 3
PROHIBITED_CONTENT = 4
IMAGE_SAFETY = 5
block_reason: "GenerateContentResponse.PromptFeedback.BlockReason" = (
proto.Field(
proto.ENUM,
number=1,
enum="GenerateContentResponse.PromptFeedback.BlockReason",
)
)
safety_ratings: MutableSequence[safety.SafetyRating] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=safety.SafetyRating,
)
[docs] class UsageMetadata(proto.Message):
r"""Metadata on the generation request's token usage.
Attributes:
prompt_token_count (int):
Number of tokens in the prompt. When ``cached_content`` is
set, this is still the total effective prompt size meaning
this includes the number of tokens in the cached content.
cached_content_token_count (int):
Number of tokens in the cached part of the
prompt (the cached content)
candidates_token_count (int):
Total number of tokens across all the
generated response candidates.
total_token_count (int):
Total token count for the generation request
(prompt + response candidates).
"""
prompt_token_count: int = proto.Field(
proto.INT32,
number=1,
)
cached_content_token_count: int = proto.Field(
proto.INT32,
number=4,
)
candidates_token_count: int = proto.Field(
proto.INT32,
number=2,
)
total_token_count: int = proto.Field(
proto.INT32,
number=3,
)
candidates: MutableSequence["Candidate"] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message="Candidate",
)
prompt_feedback: PromptFeedback = proto.Field(
proto.MESSAGE,
number=2,
message=PromptFeedback,
)
usage_metadata: UsageMetadata = proto.Field(
proto.MESSAGE,
number=3,
message=UsageMetadata,
)
model_version: str = proto.Field(
proto.STRING,
number=4,
)
[docs]class Candidate(proto.Message):
r"""A response candidate generated from the model.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
index (int):
Output only. Index of the candidate in the
list of response candidates.
This field is a member of `oneof`_ ``_index``.
content (google.ai.generativelanguage_v1alpha.types.Content):
Output only. Generated content returned from
the model.
finish_reason (google.ai.generativelanguage_v1alpha.types.Candidate.FinishReason):
Optional. Output only. The reason why the
model stopped generating tokens.
If empty, the model has not stopped generating
tokens.
safety_ratings (MutableSequence[google.ai.generativelanguage_v1alpha.types.SafetyRating]):
List of ratings for the safety of a response
candidate.
There is at most one rating per category.
citation_metadata (google.ai.generativelanguage_v1alpha.types.CitationMetadata):
Output only. Citation information for model-generated
candidate.
This field may be populated with recitation information for
any text included in the ``content``. These are passages
that are "recited" from copyrighted material in the
foundational LLM's training data.
token_count (int):
Output only. Token count for this candidate.
grounding_attributions (MutableSequence[google.ai.generativelanguage_v1alpha.types.GroundingAttribution]):
Output only. Attribution information for sources that
contributed to a grounded answer.
This field is populated for ``GenerateAnswer`` calls.
grounding_metadata (google.ai.generativelanguage_v1alpha.types.GroundingMetadata):
Output only. Grounding metadata for the candidate.
This field is populated for ``GenerateContent`` calls.
avg_logprobs (float):
Output only. Average log probability score of
the candidate.
logprobs_result (google.ai.generativelanguage_v1alpha.types.LogprobsResult):
Output only. Log-likelihood scores for the
response tokens and top tokens
"""
[docs] class FinishReason(proto.Enum):
r"""Defines the reason why the model stopped generating tokens.
Values:
FINISH_REASON_UNSPECIFIED (0):
Default value. This value is unused.
STOP (1):
Natural stop point of the model or provided
stop sequence.
MAX_TOKENS (2):
The maximum number of tokens as specified in
the request was reached.
SAFETY (3):
The response candidate content was flagged
for safety reasons.
RECITATION (4):
The response candidate content was flagged
for recitation reasons.
LANGUAGE (6):
The response candidate content was flagged
for using an unsupported language.
OTHER (5):
Unknown reason.
BLOCKLIST (7):
Token generation stopped because the content
contains forbidden terms.
PROHIBITED_CONTENT (8):
Token generation stopped for potentially
containing prohibited content.
SPII (9):
Token generation stopped because the content
potentially contains Sensitive Personally
Identifiable Information (SPII).
MALFORMED_FUNCTION_CALL (10):
The function call generated by the model is
invalid.
IMAGE_SAFETY (11):
Token generation stopped because generated
images contain safety violations.
"""
FINISH_REASON_UNSPECIFIED = 0
STOP = 1
MAX_TOKENS = 2
SAFETY = 3
RECITATION = 4
LANGUAGE = 6
OTHER = 5
BLOCKLIST = 7
PROHIBITED_CONTENT = 8
SPII = 9
MALFORMED_FUNCTION_CALL = 10
IMAGE_SAFETY = 11
index: int = proto.Field(
proto.INT32,
number=3,
optional=True,
)
content: gag_content.Content = proto.Field(
proto.MESSAGE,
number=1,
message=gag_content.Content,
)
finish_reason: FinishReason = proto.Field(
proto.ENUM,
number=2,
enum=FinishReason,
)
safety_ratings: MutableSequence[safety.SafetyRating] = proto.RepeatedField(
proto.MESSAGE,
number=5,
message=safety.SafetyRating,
)
citation_metadata: citation.CitationMetadata = proto.Field(
proto.MESSAGE,
number=6,
message=citation.CitationMetadata,
)
token_count: int = proto.Field(
proto.INT32,
number=7,
)
grounding_attributions: MutableSequence["GroundingAttribution"] = (
proto.RepeatedField(
proto.MESSAGE,
number=8,
message="GroundingAttribution",
)
)
grounding_metadata: "GroundingMetadata" = proto.Field(
proto.MESSAGE,
number=9,
message="GroundingMetadata",
)
avg_logprobs: float = proto.Field(
proto.DOUBLE,
number=10,
)
logprobs_result: "LogprobsResult" = proto.Field(
proto.MESSAGE,
number=11,
message="LogprobsResult",
)
[docs]class LogprobsResult(proto.Message):
r"""Logprobs Result
Attributes:
top_candidates (MutableSequence[google.ai.generativelanguage_v1alpha.types.LogprobsResult.TopCandidates]):
Length = total number of decoding steps.
chosen_candidates (MutableSequence[google.ai.generativelanguage_v1alpha.types.LogprobsResult.Candidate]):
Length = total number of decoding steps. The chosen
candidates may or may not be in top_candidates.
"""
[docs] class Candidate(proto.Message):
r"""Candidate for the logprobs token and score.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
token (str):
The candidate’s token string value.
This field is a member of `oneof`_ ``_token``.
token_id (int):
The candidate’s token id value.
This field is a member of `oneof`_ ``_token_id``.
log_probability (float):
The candidate's log probability.
This field is a member of `oneof`_ ``_log_probability``.
"""
token: str = proto.Field(
proto.STRING,
number=1,
optional=True,
)
token_id: int = proto.Field(
proto.INT32,
number=3,
optional=True,
)
log_probability: float = proto.Field(
proto.FLOAT,
number=2,
optional=True,
)
[docs] class TopCandidates(proto.Message):
r"""Candidates with top log probabilities at each decoding step.
Attributes:
candidates (MutableSequence[google.ai.generativelanguage_v1alpha.types.LogprobsResult.Candidate]):
Sorted by log probability in descending
order.
"""
candidates: MutableSequence["LogprobsResult.Candidate"] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message="LogprobsResult.Candidate",
)
top_candidates: MutableSequence[TopCandidates] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message=TopCandidates,
)
chosen_candidates: MutableSequence[Candidate] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=Candidate,
)
[docs]class AttributionSourceId(proto.Message):
r"""Identifier for the source contributing to this attribution.
This message has `oneof`_ fields (mutually exclusive fields).
For each oneof, at most one member field can be set at the same time.
Setting any member of the oneof automatically clears all other
members.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
grounding_passage (google.ai.generativelanguage_v1alpha.types.AttributionSourceId.GroundingPassageId):
Identifier for an inline passage.
This field is a member of `oneof`_ ``source``.
semantic_retriever_chunk (google.ai.generativelanguage_v1alpha.types.AttributionSourceId.SemanticRetrieverChunk):
Identifier for a ``Chunk`` fetched via Semantic Retriever.
This field is a member of `oneof`_ ``source``.
"""
[docs] class GroundingPassageId(proto.Message):
r"""Identifier for a part within a ``GroundingPassage``.
Attributes:
passage_id (str):
Output only. ID of the passage matching the
``GenerateAnswerRequest``'s ``GroundingPassage.id``.
part_index (int):
Output only. Index of the part within the
``GenerateAnswerRequest``'s ``GroundingPassage.content``.
"""
passage_id: str = proto.Field(
proto.STRING,
number=1,
)
part_index: int = proto.Field(
proto.INT32,
number=2,
)
[docs] class SemanticRetrieverChunk(proto.Message):
r"""Identifier for a ``Chunk`` retrieved via Semantic Retriever
specified in the ``GenerateAnswerRequest`` using
``SemanticRetrieverConfig``.
Attributes:
source (str):
Output only. Name of the source matching the request's
``SemanticRetrieverConfig.source``. Example: ``corpora/123``
or ``corpora/123/documents/abc``
chunk (str):
Output only. Name of the ``Chunk`` containing the attributed
text. Example: ``corpora/123/documents/abc/chunks/xyz``
"""
source: str = proto.Field(
proto.STRING,
number=1,
)
chunk: str = proto.Field(
proto.STRING,
number=2,
)
grounding_passage: GroundingPassageId = proto.Field(
proto.MESSAGE,
number=1,
oneof="source",
message=GroundingPassageId,
)
semantic_retriever_chunk: SemanticRetrieverChunk = proto.Field(
proto.MESSAGE,
number=2,
oneof="source",
message=SemanticRetrieverChunk,
)
[docs]class GroundingAttribution(proto.Message):
r"""Attribution for a source that contributed to an answer.
Attributes:
source_id (google.ai.generativelanguage_v1alpha.types.AttributionSourceId):
Output only. Identifier for the source
contributing to this attribution.
content (google.ai.generativelanguage_v1alpha.types.Content):
Grounding source content that makes up this
attribution.
"""
source_id: "AttributionSourceId" = proto.Field(
proto.MESSAGE,
number=3,
message="AttributionSourceId",
)
content: gag_content.Content = proto.Field(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
[docs]class RetrievalMetadata(proto.Message):
r"""Metadata related to retrieval in the grounding flow.
Attributes:
google_search_dynamic_retrieval_score (float):
Optional. Score indicating how likely information from
google search could help answer the prompt. The score is in
the range [0, 1], where 0 is the least likely and 1 is the
most likely. This score is only populated when google search
grounding and dynamic retrieval is enabled. It will be
compared to the threshold to determine whether to trigger
google search.
"""
google_search_dynamic_retrieval_score: float = proto.Field(
proto.FLOAT,
number=2,
)
[docs]class GroundingMetadata(proto.Message):
r"""Metadata returned to client when grounding is enabled.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
search_entry_point (google.ai.generativelanguage_v1alpha.types.SearchEntryPoint):
Optional. Google search entry for the
following-up web searches.
This field is a member of `oneof`_ ``_search_entry_point``.
grounding_chunks (MutableSequence[google.ai.generativelanguage_v1alpha.types.GroundingChunk]):
List of supporting references retrieved from
specified grounding source.
grounding_supports (MutableSequence[google.ai.generativelanguage_v1alpha.types.GroundingSupport]):
List of grounding support.
retrieval_metadata (google.ai.generativelanguage_v1alpha.types.RetrievalMetadata):
Metadata related to retrieval in the
grounding flow.
This field is a member of `oneof`_ ``_retrieval_metadata``.
web_search_queries (MutableSequence[str]):
Web search queries for the following-up web
search.
"""
search_entry_point: "SearchEntryPoint" = proto.Field(
proto.MESSAGE,
number=1,
optional=True,
message="SearchEntryPoint",
)
grounding_chunks: MutableSequence["GroundingChunk"] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message="GroundingChunk",
)
grounding_supports: MutableSequence["GroundingSupport"] = proto.RepeatedField(
proto.MESSAGE,
number=3,
message="GroundingSupport",
)
retrieval_metadata: "RetrievalMetadata" = proto.Field(
proto.MESSAGE,
number=4,
optional=True,
message="RetrievalMetadata",
)
web_search_queries: MutableSequence[str] = proto.RepeatedField(
proto.STRING,
number=5,
)
[docs]class SearchEntryPoint(proto.Message):
r"""Google search entry point.
Attributes:
rendered_content (str):
Optional. Web content snippet that can be
embedded in a web page or an app webview.
sdk_blob (bytes):
Optional. Base64 encoded JSON representing
array of <search term, search url> tuple.
"""
rendered_content: str = proto.Field(
proto.STRING,
number=1,
)
sdk_blob: bytes = proto.Field(
proto.BYTES,
number=2,
)
[docs]class GroundingChunk(proto.Message):
r"""Grounding chunk.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
web (google.ai.generativelanguage_v1alpha.types.GroundingChunk.Web):
Grounding chunk from the web.
This field is a member of `oneof`_ ``chunk_type``.
"""
[docs] class Web(proto.Message):
r"""Chunk from the web.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
uri (str):
URI reference of the chunk.
This field is a member of `oneof`_ ``_uri``.
title (str):
Title of the chunk.
This field is a member of `oneof`_ ``_title``.
"""
uri: str = proto.Field(
proto.STRING,
number=1,
optional=True,
)
title: str = proto.Field(
proto.STRING,
number=2,
optional=True,
)
web: Web = proto.Field(
proto.MESSAGE,
number=1,
oneof="chunk_type",
message=Web,
)
[docs]class Segment(proto.Message):
r"""Segment of the content.
Attributes:
part_index (int):
Output only. The index of a Part object
within its parent Content object.
start_index (int):
Output only. Start index in the given Part,
measured in bytes. Offset from the start of the
Part, inclusive, starting at zero.
end_index (int):
Output only. End index in the given Part,
measured in bytes. Offset from the start of the
Part, exclusive, starting at zero.
text (str):
Output only. The text corresponding to the
segment from the response.
"""
part_index: int = proto.Field(
proto.INT32,
number=1,
)
start_index: int = proto.Field(
proto.INT32,
number=2,
)
end_index: int = proto.Field(
proto.INT32,
number=3,
)
text: str = proto.Field(
proto.STRING,
number=4,
)
[docs]class GroundingSupport(proto.Message):
r"""Grounding support.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
segment (google.ai.generativelanguage_v1alpha.types.Segment):
Segment of the content this support belongs
to.
This field is a member of `oneof`_ ``_segment``.
grounding_chunk_indices (MutableSequence[int]):
A list of indices (into 'grounding_chunk') specifying the
citations associated with the claim. For instance [1,3,4]
means that grounding_chunk[1], grounding_chunk[3],
grounding_chunk[4] are the retrieved content attributed to
the claim.
confidence_scores (MutableSequence[float]):
Confidence score of the support references. Ranges from 0 to
1. 1 is the most confident. This list must have the same
size as the grounding_chunk_indices.
"""
segment: "Segment" = proto.Field(
proto.MESSAGE,
number=1,
optional=True,
message="Segment",
)
grounding_chunk_indices: MutableSequence[int] = proto.RepeatedField(
proto.INT32,
number=2,
)
confidence_scores: MutableSequence[float] = proto.RepeatedField(
proto.FLOAT,
number=3,
)
[docs]class GenerateAnswerRequest(proto.Message):
r"""Request to generate a grounded answer from the ``Model``.
This message has `oneof`_ fields (mutually exclusive fields).
For each oneof, at most one member field can be set at the same time.
Setting any member of the oneof automatically clears all other
members.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
inline_passages (google.ai.generativelanguage_v1alpha.types.GroundingPassages):
Passages provided inline with the request.
This field is a member of `oneof`_ ``grounding_source``.
semantic_retriever (google.ai.generativelanguage_v1alpha.types.SemanticRetrieverConfig):
Content retrieved from resources created via
the Semantic Retriever API.
This field is a member of `oneof`_ ``grounding_source``.
model (str):
Required. The name of the ``Model`` to use for generating
the grounded response.
Format: ``model=models/{model}``.
contents (MutableSequence[google.ai.generativelanguage_v1alpha.types.Content]):
Required. The content of the current conversation with the
``Model``. For single-turn queries, this is a single
question to answer. For multi-turn queries, this is a
repeated field that contains conversation history and the
last ``Content`` in the list containing the question.
Note: ``GenerateAnswer`` only supports queries in English.
answer_style (google.ai.generativelanguage_v1alpha.types.GenerateAnswerRequest.AnswerStyle):
Required. Style in which answers should be
returned.
safety_settings (MutableSequence[google.ai.generativelanguage_v1alpha.types.SafetySetting]):
Optional. A list of unique ``SafetySetting`` instances for
blocking unsafe content.
This will be enforced on the
``GenerateAnswerRequest.contents`` and
``GenerateAnswerResponse.candidate``. There should not be
more than one setting for each ``SafetyCategory`` type. The
API will block any contents and responses that fail to meet
the thresholds set by these settings. This list overrides
the default settings for each ``SafetyCategory`` specified
in the safety_settings. If there is no ``SafetySetting`` for
a given ``SafetyCategory`` provided in the list, the API
will use the default safety setting for that category. Harm
categories HARM_CATEGORY_HATE_SPEECH,
HARM_CATEGORY_SEXUALLY_EXPLICIT,
HARM_CATEGORY_DANGEROUS_CONTENT, HARM_CATEGORY_HARASSMENT
are supported. Refer to the
`guide <https://ai.google.dev/gemini-api/docs/safety-settings>`__
for detailed information on available safety settings. Also
refer to the `Safety
guidance <https://ai.google.dev/gemini-api/docs/safety-guidance>`__
to learn how to incorporate safety considerations in your AI
applications.
temperature (float):
Optional. Controls the randomness of the output.
Values can range from [0.0,1.0], inclusive. A value closer
to 1.0 will produce responses that are more varied and
creative, while a value closer to 0.0 will typically result
in more straightforward responses from the model. A low
temperature (~0.2) is usually recommended for
Attributed-Question-Answering use cases.
This field is a member of `oneof`_ ``_temperature``.
"""
[docs] class AnswerStyle(proto.Enum):
r"""Style for grounded answers.
Values:
ANSWER_STYLE_UNSPECIFIED (0):
Unspecified answer style.
ABSTRACTIVE (1):
Succint but abstract style.
EXTRACTIVE (2):
Very brief and extractive style.
VERBOSE (3):
Verbose style including extra details. The
response may be formatted as a sentence,
paragraph, multiple paragraphs, or bullet
points, etc.
"""
ANSWER_STYLE_UNSPECIFIED = 0
ABSTRACTIVE = 1
EXTRACTIVE = 2
VERBOSE = 3
inline_passages: gag_content.GroundingPassages = proto.Field(
proto.MESSAGE,
number=6,
oneof="grounding_source",
message=gag_content.GroundingPassages,
)
semantic_retriever: "SemanticRetrieverConfig" = proto.Field(
proto.MESSAGE,
number=7,
oneof="grounding_source",
message="SemanticRetrieverConfig",
)
model: str = proto.Field(
proto.STRING,
number=1,
)
contents: MutableSequence[gag_content.Content] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
answer_style: AnswerStyle = proto.Field(
proto.ENUM,
number=5,
enum=AnswerStyle,
)
safety_settings: MutableSequence[safety.SafetySetting] = proto.RepeatedField(
proto.MESSAGE,
number=3,
message=safety.SafetySetting,
)
temperature: float = proto.Field(
proto.FLOAT,
number=4,
optional=True,
)
[docs]class GenerateAnswerResponse(proto.Message):
r"""Response from the model for a grounded answer.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
answer (google.ai.generativelanguage_v1alpha.types.Candidate):
Candidate answer from the model.
Note: The model *always* attempts to provide a grounded
answer, even when the answer is unlikely to be answerable
from the given passages. In that case, a low-quality or
ungrounded answer may be provided, along with a low
``answerable_probability``.
answerable_probability (float):
Output only. The model's estimate of the probability that
its answer is correct and grounded in the input passages.
A low ``answerable_probability`` indicates that the answer
might not be grounded in the sources.
When ``answerable_probability`` is low, you may want to:
- Display a message to the effect of "We couldn’t answer
that question" to the user.
- Fall back to a general-purpose LLM that answers the
question from world knowledge. The threshold and nature of
such fallbacks will depend on individual use cases.
``0.5`` is a good starting threshold.
This field is a member of `oneof`_ ``_answerable_probability``.
input_feedback (google.ai.generativelanguage_v1alpha.types.GenerateAnswerResponse.InputFeedback):
Output only. Feedback related to the input data used to
answer the question, as opposed to the model-generated
response to the question.
The input data can be one or more of the following:
- Question specified by the last entry in
``GenerateAnswerRequest.content``
- Conversation history specified by the other entries in
``GenerateAnswerRequest.content``
- Grounding sources
(``GenerateAnswerRequest.semantic_retriever`` or
``GenerateAnswerRequest.inline_passages``)
This field is a member of `oneof`_ ``_input_feedback``.
"""
[docs] class InputFeedback(proto.Message):
r"""Feedback related to the input data used to answer the
question, as opposed to the model-generated response to the
question.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
block_reason (google.ai.generativelanguage_v1alpha.types.GenerateAnswerResponse.InputFeedback.BlockReason):
Optional. If set, the input was blocked and
no candidates are returned. Rephrase the input.
This field is a member of `oneof`_ ``_block_reason``.
safety_ratings (MutableSequence[google.ai.generativelanguage_v1alpha.types.SafetyRating]):
Ratings for safety of the input.
There is at most one rating per category.
"""
[docs] class BlockReason(proto.Enum):
r"""Specifies what was the reason why input was blocked.
Values:
BLOCK_REASON_UNSPECIFIED (0):
Default value. This value is unused.
SAFETY (1):
Input was blocked due to safety reasons. Inspect
``safety_ratings`` to understand which safety category
blocked it.
OTHER (2):
Input was blocked due to other reasons.
"""
BLOCK_REASON_UNSPECIFIED = 0
SAFETY = 1
OTHER = 2
block_reason: "GenerateAnswerResponse.InputFeedback.BlockReason" = proto.Field(
proto.ENUM,
number=1,
optional=True,
enum="GenerateAnswerResponse.InputFeedback.BlockReason",
)
safety_ratings: MutableSequence[safety.SafetyRating] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=safety.SafetyRating,
)
answer: "Candidate" = proto.Field(
proto.MESSAGE,
number=1,
message="Candidate",
)
answerable_probability: float = proto.Field(
proto.FLOAT,
number=2,
optional=True,
)
input_feedback: InputFeedback = proto.Field(
proto.MESSAGE,
number=3,
optional=True,
message=InputFeedback,
)
[docs]class EmbedContentRequest(proto.Message):
r"""Request containing the ``Content`` for the model to embed.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
model (str):
Required. The model's resource name. This serves as an ID
for the Model to use.
This name should match a model name returned by the
``ListModels`` method.
Format: ``models/{model}``
content (google.ai.generativelanguage_v1alpha.types.Content):
Required. The content to embed. Only the ``parts.text``
fields will be counted.
task_type (google.ai.generativelanguage_v1alpha.types.TaskType):
Optional. Optional task type for which the embeddings will
be used. Can only be set for ``models/embedding-001``.
This field is a member of `oneof`_ ``_task_type``.
title (str):
Optional. An optional title for the text. Only applicable
when TaskType is ``RETRIEVAL_DOCUMENT``.
Note: Specifying a ``title`` for ``RETRIEVAL_DOCUMENT``
provides better quality embeddings for retrieval.
This field is a member of `oneof`_ ``_title``.
output_dimensionality (int):
Optional. Optional reduced dimension for the output
embedding. If set, excessive values in the output embedding
are truncated from the end. Supported by newer models since
2024 only. You cannot set this value if using the earlier
model (``models/embedding-001``).
This field is a member of `oneof`_ ``_output_dimensionality``.
"""
model: str = proto.Field(
proto.STRING,
number=1,
)
content: gag_content.Content = proto.Field(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
task_type: "TaskType" = proto.Field(
proto.ENUM,
number=3,
optional=True,
enum="TaskType",
)
title: str = proto.Field(
proto.STRING,
number=4,
optional=True,
)
output_dimensionality: int = proto.Field(
proto.INT32,
number=5,
optional=True,
)
[docs]class ContentEmbedding(proto.Message):
r"""A list of floats representing an embedding.
Attributes:
values (MutableSequence[float]):
The embedding values.
"""
values: MutableSequence[float] = proto.RepeatedField(
proto.FLOAT,
number=1,
)
[docs]class EmbedContentResponse(proto.Message):
r"""The response to an ``EmbedContentRequest``.
Attributes:
embedding (google.ai.generativelanguage_v1alpha.types.ContentEmbedding):
Output only. The embedding generated from the
input content.
"""
embedding: "ContentEmbedding" = proto.Field(
proto.MESSAGE,
number=1,
message="ContentEmbedding",
)
[docs]class BatchEmbedContentsRequest(proto.Message):
r"""Batch request to get embeddings from the model for a list of
prompts.
Attributes:
model (str):
Required. The model's resource name. This serves as an ID
for the Model to use.
This name should match a model name returned by the
``ListModels`` method.
Format: ``models/{model}``
requests (MutableSequence[google.ai.generativelanguage_v1alpha.types.EmbedContentRequest]):
Required. Embed requests for the batch. The model in each of
these requests must match the model specified
``BatchEmbedContentsRequest.model``.
"""
model: str = proto.Field(
proto.STRING,
number=1,
)
requests: MutableSequence["EmbedContentRequest"] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message="EmbedContentRequest",
)
[docs]class BatchEmbedContentsResponse(proto.Message):
r"""The response to a ``BatchEmbedContentsRequest``.
Attributes:
embeddings (MutableSequence[google.ai.generativelanguage_v1alpha.types.ContentEmbedding]):
Output only. The embeddings for each request,
in the same order as provided in the batch
request.
"""
embeddings: MutableSequence["ContentEmbedding"] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message="ContentEmbedding",
)
[docs]class CountTokensRequest(proto.Message):
r"""Counts the number of tokens in the ``prompt`` sent to a model.
Models may tokenize text differently, so each model may return a
different ``token_count``.
Attributes:
model (str):
Required. The model's resource name. This serves as an ID
for the Model to use.
This name should match a model name returned by the
``ListModels`` method.
Format: ``models/{model}``
contents (MutableSequence[google.ai.generativelanguage_v1alpha.types.Content]):
Optional. The input given to the model as a prompt. This
field is ignored when ``generate_content_request`` is set.
generate_content_request (google.ai.generativelanguage_v1alpha.types.GenerateContentRequest):
Optional. The overall input given to the ``Model``. This
includes the prompt as well as other model steering
information like `system
instructions <https://ai.google.dev/gemini-api/docs/system-instructions>`__,
and/or function declarations for `function
calling <https://ai.google.dev/gemini-api/docs/function-calling>`__.
``Model``\ s/``Content``\ s and
``generate_content_request``\ s are mutually exclusive. You
can either send ``Model`` + ``Content``\ s or a
``generate_content_request``, but never both.
"""
model: str = proto.Field(
proto.STRING,
number=1,
)
contents: MutableSequence[gag_content.Content] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=gag_content.Content,
)
generate_content_request: "GenerateContentRequest" = proto.Field(
proto.MESSAGE,
number=3,
message="GenerateContentRequest",
)
[docs]class CountTokensResponse(proto.Message):
r"""A response from ``CountTokens``.
It returns the model's ``token_count`` for the ``prompt``.
Attributes:
total_tokens (int):
The number of tokens that the ``Model`` tokenizes the
``prompt`` into. Always non-negative.
cached_content_token_count (int):
Number of tokens in the cached part of the
prompt (the cached content).
"""
total_tokens: int = proto.Field(
proto.INT32,
number=1,
)
cached_content_token_count: int = proto.Field(
proto.INT32,
number=5,
)
[docs]class BidiGenerateContentSetup(proto.Message):
r"""Message to be sent in the first and only first
``BidiGenerateContentClientMessage``. Contains configuration that
will apply for the duration of the streaming RPC.
Clients should wait for a ``BidiGenerateContentSetupComplete``
message before sending any additional messages.
Attributes:
model (str):
Required. The model's resource name. This serves as an ID
for the Model to use.
Format: ``models/{model}``
generation_config (google.ai.generativelanguage_v1alpha.types.GenerationConfig):
Optional. Generation config.
The following fields are not supported:
- ``response_logprobs``
- ``response_mime_type``
- ``logprobs``
- ``response_schema``
- ``stop_sequence``
- ``routing_config``
- ``audio_timestamp``
system_instruction (google.ai.generativelanguage_v1alpha.types.Content):
Optional. The user provided system
instructions for the model.
Note: Only text should be used in parts and
content in each part will be in a separate
paragraph.
tools (MutableSequence[google.ai.generativelanguage_v1alpha.types.Tool]):
Optional. A list of ``Tools`` the model may use to generate
the next response.
A ``Tool`` is a piece of code that enables the system to
interact with external systems to perform an action, or set
of actions, outside of knowledge and scope of the model.
"""
model: str = proto.Field(
proto.STRING,
number=1,
)
generation_config: "GenerationConfig" = proto.Field(
proto.MESSAGE,
number=2,
message="GenerationConfig",
)
system_instruction: gag_content.Content = proto.Field(
proto.MESSAGE,
number=3,
message=gag_content.Content,
)
tools: MutableSequence[gag_content.Tool] = proto.RepeatedField(
proto.MESSAGE,
number=4,
message=gag_content.Tool,
)
[docs]class BidiGenerateContentClientContent(proto.Message):
r"""Incremental update of the current conversation delivered from
the client. All of the content here is unconditionally appended
to the conversation history and used as part of the prompt to
the model to generate content.
A message here will interrupt any current model generation.
Attributes:
turns (MutableSequence[google.ai.generativelanguage_v1alpha.types.Content]):
Optional. The content appended to the current
conversation with the model.
For single-turn queries, this is a single
instance. For multi-turn queries, this is a
repeated field that contains conversation
history and the latest request.
turn_complete (bool):
Optional. If true, indicates that the server
content generation should start with the
currently accumulated prompt. Otherwise, the
server awaits additional messages before
starting generation.
"""
turns: MutableSequence[gag_content.Content] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message=gag_content.Content,
)
turn_complete: bool = proto.Field(
proto.BOOL,
number=2,
)
[docs]class BidiGenerateContentRealtimeInput(proto.Message):
r"""User input that is sent in real time.
This is different from
[BidiGenerateContentClientContent][google.ai.generativelanguage.v1alpha.BidiGenerateContentClientContent]
in a few ways:
- Can be sent continuously without interruption to model generation.
- If there is a need to mix data interleaved across the
[BidiGenerateContentClientContent][google.ai.generativelanguage.v1alpha.BidiGenerateContentClientContent]
and the
[BidiGenerateContentRealtimeInput][google.ai.generativelanguage.v1alpha.BidiGenerateContentRealtimeInput],
the server attempts to optimize for best response, but there are
no guarantees.
- End of turn is not explicitly specified, but is rather derived
from user activity (for example, end of speech).
- Even before the end of turn, the data is processed incrementally
to optimize for a fast start of the response from the model.
- Is always direct user input that is sent in real time. Can be sent
continuously without interruptions. The model automatically
detects the beginning and the end of user speech and starts or
terminates streaming the response accordingly. Data is processed
incrementally as it arrives, minimizing latency.
Attributes:
media_chunks (MutableSequence[google.ai.generativelanguage_v1alpha.types.Blob]):
Optional. Inlined bytes data for media input.
"""
media_chunks: MutableSequence[gag_content.Blob] = proto.RepeatedField(
proto.MESSAGE,
number=1,
message=gag_content.Blob,
)
[docs]class BidiGenerateContentToolResponse(proto.Message):
r"""Client generated response to a ``ToolCall`` received from the
server. Individual ``FunctionResponse`` objects are matched to the
respective ``FunctionCall`` objects by the ``id`` field.
Note that in the unary and server-streaming GenerateContent APIs
function calling happens by exchanging the ``Content`` parts, while
in the bidi GenerateContent APIs function calling happens over these
dedicated set of messages.
Attributes:
function_responses (MutableSequence[google.ai.generativelanguage_v1alpha.types.FunctionResponse]):
Optional. The response to the function calls.
"""
function_responses: MutableSequence[gag_content.FunctionResponse] = (
proto.RepeatedField(
proto.MESSAGE,
number=1,
message=gag_content.FunctionResponse,
)
)
[docs]class BidiGenerateContentClientMessage(proto.Message):
r"""Messages sent by the client in the BidiGenerateContent call.
This message has `oneof`_ fields (mutually exclusive fields).
For each oneof, at most one member field can be set at the same time.
Setting any member of the oneof automatically clears all other
members.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
setup (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentSetup):
Optional. Session configuration sent in the
first and only first client message.
This field is a member of `oneof`_ ``message_type``.
client_content (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentClientContent):
Optional. Incremental update of the current
conversation delivered from the client.
This field is a member of `oneof`_ ``message_type``.
realtime_input (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentRealtimeInput):
Optional. User input that is sent in real
time.
This field is a member of `oneof`_ ``message_type``.
tool_response (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentToolResponse):
Optional. Response to a ``ToolCallMessage`` received from
the server.
This field is a member of `oneof`_ ``message_type``.
"""
setup: "BidiGenerateContentSetup" = proto.Field(
proto.MESSAGE,
number=1,
oneof="message_type",
message="BidiGenerateContentSetup",
)
client_content: "BidiGenerateContentClientContent" = proto.Field(
proto.MESSAGE,
number=2,
oneof="message_type",
message="BidiGenerateContentClientContent",
)
realtime_input: "BidiGenerateContentRealtimeInput" = proto.Field(
proto.MESSAGE,
number=3,
oneof="message_type",
message="BidiGenerateContentRealtimeInput",
)
tool_response: "BidiGenerateContentToolResponse" = proto.Field(
proto.MESSAGE,
number=4,
oneof="message_type",
message="BidiGenerateContentToolResponse",
)
[docs]class BidiGenerateContentSetupComplete(proto.Message):
r"""Sent in response to a ``BidiGenerateContentSetup`` message from the
client.
"""
[docs]class BidiGenerateContentServerContent(proto.Message):
r"""Incremental server update generated by the model in response
to client messages.
Content is generated as quickly as possible, and not in real
time. Clients may choose to buffer and play it out in real time.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
model_turn (google.ai.generativelanguage_v1alpha.types.Content):
Output only. The content that the model has
generated as part of the current conversation
with the user.
This field is a member of `oneof`_ ``_model_turn``.
turn_complete (bool):
Output only. If true, indicates that the model is done
generating. Generation will only start in response to
additional client messages. Can be set alongside
``content``, indicating that the ``content`` is the last in
the turn.
interrupted (bool):
Output only. If true, indicates that a client
message has interrupted current model
generation. If the client is playing out the
content in real time, this is a good signal to
stop and empty the current playback queue.
grounding_metadata (google.ai.generativelanguage_v1alpha.types.GroundingMetadata):
Output only. Grounding metadata for the
generated content.
"""
model_turn: gag_content.Content = proto.Field(
proto.MESSAGE,
number=1,
optional=True,
message=gag_content.Content,
)
turn_complete: bool = proto.Field(
proto.BOOL,
number=2,
)
interrupted: bool = proto.Field(
proto.BOOL,
number=3,
)
grounding_metadata: "GroundingMetadata" = proto.Field(
proto.MESSAGE,
number=4,
message="GroundingMetadata",
)
[docs]class BidiGenerateContentToolCall(proto.Message):
r"""Request for the client to execute the ``function_calls`` and return
the responses with the matching ``id``\ s.
Attributes:
function_calls (MutableSequence[google.ai.generativelanguage_v1alpha.types.FunctionCall]):
Output only. The function call to be
executed.
"""
function_calls: MutableSequence[gag_content.FunctionCall] = proto.RepeatedField(
proto.MESSAGE,
number=2,
message=gag_content.FunctionCall,
)
[docs]class BidiGenerateContentToolCallCancellation(proto.Message):
r"""Notification for the client that a previously issued
``ToolCallMessage`` with the specified ``id``\ s should have been
not executed and should be cancelled. If there were side-effects to
those tool calls, clients may attempt to undo the tool calls. This
message occurs only in cases where the clients interrupt server
turns.
Attributes:
ids (MutableSequence[str]):
Output only. The ids of the tool calls to be
cancelled.
"""
ids: MutableSequence[str] = proto.RepeatedField(
proto.STRING,
number=1,
)
[docs]class BidiGenerateContentServerMessage(proto.Message):
r"""Response message for the BidiGenerateContent call.
This message has `oneof`_ fields (mutually exclusive fields).
For each oneof, at most one member field can be set at the same time.
Setting any member of the oneof automatically clears all other
members.
.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields
Attributes:
setup_complete (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentSetupComplete):
Output only. Sent in response to a
``BidiGenerateContentSetup`` message from the client when
setup is complete.
This field is a member of `oneof`_ ``message_type``.
server_content (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentServerContent):
Output only. Content generated by the model
in response to client messages.
This field is a member of `oneof`_ ``message_type``.
tool_call (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentToolCall):
Output only. Request for the client to execute the
``function_calls`` and return the responses with the
matching ``id``\ s.
This field is a member of `oneof`_ ``message_type``.
tool_call_cancellation (google.ai.generativelanguage_v1alpha.types.BidiGenerateContentToolCallCancellation):
Output only. Notification for the client that a previously
issued ``ToolCallMessage`` with the specified ``id``\ s
should be cancelled.
This field is a member of `oneof`_ ``message_type``.
"""
setup_complete: "BidiGenerateContentSetupComplete" = proto.Field(
proto.MESSAGE,
number=2,
oneof="message_type",
message="BidiGenerateContentSetupComplete",
)
server_content: "BidiGenerateContentServerContent" = proto.Field(
proto.MESSAGE,
number=3,
oneof="message_type",
message="BidiGenerateContentServerContent",
)
tool_call: "BidiGenerateContentToolCall" = proto.Field(
proto.MESSAGE,
number=4,
oneof="message_type",
message="BidiGenerateContentToolCall",
)
tool_call_cancellation: "BidiGenerateContentToolCallCancellation" = proto.Field(
proto.MESSAGE,
number=5,
oneof="message_type",
message="BidiGenerateContentToolCallCancellation",
)
__all__ = tuple(sorted(__protobuf__.manifest))
