Module llm (1.17.0)

LLM models.

Classes

Claude3TextGenerator

Claude3TextGenerator(
    *,
    model_name: typing.Literal[
        "claude-3-sonnet", "claude-3-haiku", "claude-3-5-sonnet", "claude-3-opus"
    ] = "claude-3-sonnet",
    session: typing.Optional[bigframes.session.Session] = None,
    connection_name: typing.Optional[str] = None
)

Claude3 text generator LLM model.

Go to Google Cloud Console -> Vertex AI -> Model Garden page to enabe the models before use. Must have the Consumer Procurement Entitlement Manager Identity and Access Management (IAM) role to enable the models. https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-partner-models#grant-permissions

The models only availabe in specific regions. Check https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-claude#regions for details.

Parameters
Name Description
model_name str, Default to "claude-3-sonnet"

The model for natural language tasks. Possible values are "claude-3-sonnet", "claude-3-haiku", "claude-3-5-sonnet" and "claude-3-opus". "claude-3-sonnet" is Anthropic's dependable combination of skills and speed. It is engineered to be dependable for scaled AI deployments across a variety of use cases. "claude-3-haiku" is Anthropic's fastest, most compact vision and text model for near-instant responses to simple queries, meant for seamless AI experiences mimicking human interactions. "claude-3-5-sonnet" is Anthropic's most powerful AI model and maintains the speed and cost of Claude 3 Sonnet, which is a mid-tier model. "claude-3-opus" is Anthropic's second-most powerful AI model, with strong performance on highly complex tasks. https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-claude#available-claude-models Default to "claude-3-sonnet".

session bigframes.Session or None

BQ session to create the model. If None, use the global default session.

connection_name str or None

Connection to connect with remote service. str of the format <PROJECT_NUMBER/PROJECT_ID>.

GeminiTextGenerator

GeminiTextGenerator(
    *,
    model_name: typing.Literal[
        "gemini-pro",
        "gemini-1.5-pro-preview-0514",
        "gemini-1.5-flash-preview-0514",
        "gemini-1.5-pro-001",
        "gemini-1.5-flash-001",
    ] = "gemini-pro",
    session: typing.Optional[bigframes.session.Session] = None,
    connection_name: typing.Optional[str] = None,
    max_iterations: int = 300
)

Gemini text generator LLM model.

Parameters
Name Description
model_name str, Default to "gemini-pro"

The model for natural language tasks. Accepted values are "gemini-pro", "gemini-1.5-pro-preview-0514", "gemini-1.5-flash-preview-0514", "gemini-1.5-pro-001" and "gemini-1.5-flash-001". Default to "gemini-pro".

session bigframes.Session or None

BQ session to create the model. If None, use the global default session.

connection_name str or None

Connection to connect with remote service. str of the format <PROJECT_NUMBER/PROJECT_ID>.

max_iterations Optional[int], Default to 300

The number of steps to run when performing supervised tuning.

PaLM2TextEmbeddingGenerator

PaLM2TextEmbeddingGenerator(
    *,
    model_name: typing.Literal[
        "textembedding-gecko", "textembedding-gecko-multilingual"
    ] = "textembedding-gecko",
    version: typing.Optional[str] = None,
    session: typing.Optional[bigframes.session.Session] = None,
    connection_name: typing.Optional[str] = None
)

PaLM2 text embedding generator LLM model.

Parameters
Name Description
model_name str, Default to "textembedding-gecko"

The model for text embedding. “textembedding-gecko” returns model embeddings for text inputs. "textembedding-gecko-multilingual" returns model embeddings for text inputs which support over 100 languages. Default to "textembedding-gecko".

version str or None

Model version. Accepted values are "001", "002", "003", "latest" etc. Will use the default version if unset. See https://cloud.google.com/vertex-ai/docs/generative-ai/learn/model-versioning for details.

session bigframes.Session or None

BQ session to create the model. If None, use the global default session.

connection_name str or None

Connection to connect with remote service. str of the format <PROJECT_NUMBER/PROJECT_ID>.

PaLM2TextGenerator

PaLM2TextGenerator(
    *,
    model_name: typing.Literal["text-bison", "text-bison-32k"] = "text-bison",
    session: typing.Optional[bigframes.session.Session] = None,
    connection_name: typing.Optional[str] = None,
    max_iterations: int = 300
)

PaLM2 text generator LLM model.

Parameters
Name Description
model_name str, Default to "text-bison"

The model for natural language tasks. “text-bison” returns model fine-tuned to follow natural language instructions and is suitable for a variety of language tasks. "text-bison-32k" supports up to 32k tokens per request. Default to "text-bison".

session bigframes.Session or None

BQ session to create the model. If None, use the global default session.

connection_name str or None

Connection to connect with remote service. str of the format <PROJECT_NUMBER/PROJECT_ID>.

max_iterations Optional[int], Default to 300

The number of steps to run when performing supervised tuning.

TextEmbeddingGenerator

TextEmbeddingGenerator(
    *,
    model_name: typing.Literal[
        "text-embedding-004", "text-multilingual-embedding-002"
    ] = "text-embedding-004",
    session: typing.Optional[bigframes.session.Session] = None,
    connection_name: typing.Optional[str] = None
)

Text embedding generator LLM model.

Parameters
Name Description
model_name str, Default to "text-embedding-004"

The model for text embedding. Possible values are "text-embedding-004" or "text-multilingual-embedding-002". text-embedding models returns model embeddings for text inputs. text-multilingual-embedding models returns model embeddings for text inputs which support over 100 languages. Default to "text-embedding-004".

session bigframes.Session or None

BQ session to create the model. If None, use the global default session.

connection_name str or None

Connection to connect with remote service. str of the format <PROJECT_NUMBER/PROJECT_ID>.