AI Client

Artificial intelligence client resource.

Allows integration with AI providers compatible with the OpenAI API, supporting chat, streaming, embeddings and MCP (Model Context Protocol) tools.

const client = _ai.client('openai')
client.model('gpt-4o')

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const result = client.chat(messages)
_out.json(result)

chat

chat(model: string, messages: Values) : Values

Description

Runs a conversation explicitly specifying the model to use, overriding the default configured model.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const response = client.chat('gpt-4o-mini', messages)
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.

Return

( Values )

Object with the full API response.

chat(model: string, messages: Values, options: Values) : Values

Description

Runs a conversation explicitly specifying the model to use, with additional options, overriding the default configured model.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

const response = client.chat('gpt-4o-mini', messages, options)
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.

Return

( Values )

Object with the full API response.

chat(model: string, messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

Runs a conversation explicitly specifying the model to use, with additional options and MCP tool support via callback, overriding the default configured model.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

const response = client.chat('gpt-4o-mini', messages, options, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( Values )

Object with the full API response.

chat(model: string, messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

Runs a conversation explicitly specifying the model to use, with MCP tool support via callback, overriding the default configured model.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'What time is it?'))

const response = client.chat('gpt-4o-mini', messages, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( Values )

Object with the full API response.

chat(messages: Values) : Values

Description

Runs a conversation with the configured AI model, sending a list of messages and returning the full response.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'system').set('content', 'You are a helpful assistant.'))
    .add(_val.map().set('role', 'user').set('content', 'What is the capital of Portugal?'))

const response = client.chat(messages)
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages. Each message must have the fields `role` (system, user, assistant) and `content`.

Return

( Values )

Object with the full API response, including choices, usage and other metadata.

chat(messages: Values, options: Values) : Values

Description

Runs a conversation with the configured AI model, with additional options such as temperature and max_tokens.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map()
    .set('temperature', 0.7)
    .set('max_tokens', 200)

const response = client.chat(messages, options)
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.

Return

( Values )

Object with the full API response.

chat(messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

Runs a conversation with the configured AI model, with additional options and MCP tool support via callback.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

const response = client.chat(messages, options, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( Values )

Object with the full API response.

chat(messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

Runs a conversation with the configured AI model with MCP tool support via callback. The callback is invoked before each tool call, allowing you to intercept or override the result.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'What time is it?'))

const response = client.chat(messages, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null // null = let the client execute normally
})
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( Values )

Object with the full API response.

embeddings

embeddings(input: string) : Values

Description

Generates a vector embedding for a text input using the configured model.

How To Use

const result = client.embeddings('The sky is blue.')
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
input	string	Text input for which the embedding will be generated.

Return

( Values )

Object with the API response, including the generated vectors and usage metadata.

embeddings(model: string, input: string) : Values

Description

Generates a vector embedding for a text input by explicitly specifying the model to use.

How To Use

const result = client.embeddings('text-embedding-3-small', 'The sky is blue.')
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the embeddings model to use, for example: `text-embedding-3-small`.
input	string	Text input for which the embedding will be generated.

Return

( Values )

Object with the API response, including the generated vectors and usage metadata.

embeddings(model: string, input: string, options: Values) : Values

Description

Generates a vector embedding for a text input by explicitly specifying the model and additional options.

How To Use

const options = _val.map().set('dimensions', 512)

const result = client.embeddings('text-embedding-3-small', 'The sky is blue.', options)
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the embeddings model to use, for example: `text-embedding-3-small`.
input	string	Text input for which the embedding will be generated.
options	Values	Additional options: `dimensions` (number of vector dimensions), `encoding_format` (`float` or `base64`), `user` (end-user identifier).

Return

( Values )

Object with the API response, including the generated vectors and usage metadata.

embeddings(model: string, inputs: Values) : Values

Description

Generates vector embeddings for multiple text inputs by explicitly specifying the model to use. The list must contain text values only.

How To Use

const texts = _val.list()
    .add('The sky is blue.')
    .add('The grass is green.')

const result = client.embeddings('text-embedding-3-small', texts)
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the embeddings model to use, for example: `text-embedding-3-small`.
inputs	Values	List of text inputs. Each element must be a plain text string.

Return

( Values )

Object with the API response, including the generated vectors for each text and usage metadata.

embeddings(model: string, inputs: Values, options: Values) : Values

Description

Generates vector embeddings for multiple text inputs by explicitly specifying the model to use and additional options. The list must contain text values only.

How To Use

const texts = _val.list()
    .add('The sky is blue.')
    .add('The grass is green.')

const options = _val.map().set('dimensions', 512)

const result = client.embeddings('text-embedding-3-small', texts, options)
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the embeddings model to use, for example: `text-embedding-3-small`.
inputs	Values	List of text inputs. Each element must be a plain text string.
options	Values	Additional options: `dimensions` (number of vector dimensions), `encoding_format` (`float` or `base64`), `user` (end-user identifier).

Return

( Values )

Object with the API response, including the generated vectors for each text and usage metadata.

embeddings(inputs: Values) : Values

Description

Generates vector embeddings for multiple text inputs using the configured model. The list must contain text values only.

How To Use

const texts = _val.list()
    .add('The sky is blue.')
    .add('The grass is green.')

const result = client.embeddings(texts)
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
inputs	Values	List of text inputs. Each element must be a plain text string.

Return

( Values )

Object with the API response, including the generated vectors for each text and usage metadata.

embeddings(inputs: Values, options: Values) : Values

Description

Generates vector embeddings for multiple text inputs using the configured model, with additional options. The list must contain text values only.

How To Use

const texts = _val.list()
    .add('The sky is blue.')
    .add('The grass is green.')

const options = _val.map().set('dimensions', 512)

const result = client.embeddings(texts, options)
_out.json(result.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
inputs	Values	List of text inputs. Each element must be a plain text string.
options	Values	Additional options: `dimensions` (number of vector dimensions), `encoding_format` (`float` or `base64`), `user` (end-user identifier).

Return

( Values )

Object with the API response, including the generated vectors for each text and usage metadata.

getMaxToolLoops

getMaxToolLoops() : int

Description

Gets the configured maximum number of tool call loops.

How To Use

const maxLoops = client.getMaxToolLoops()
_out.print(maxLoops)

Return

( int )

Maximum number of tool loops.

instance

instance() : com.openai.client.OpenAIClient

Description

Gets the internal OpenAI client instance for advanced direct use with the underlying library.

How To Use

const openAIClient = client.instance()

Return

( com.openai.client.OpenAIClient )

OpenAI client instance.

isInitialized

isInitialized() : boolean

Description

Checks whether the AI client was successfully initialized for the configured provider.

How To Use

if (!client.isInitialized()) {
    _log.error('Client not initialized.')
}

Return

( boolean )

True if the client is initialized.

maxToolLoops

maxToolLoops(maxLoops: int) : boolean

Description

Sets the maximum number of tool call cycles (tool loops) during a conversation. Prevents infinite loops when the model keeps invoking tools successively.

How To Use

client.maxToolLoops(5)

Attributes

NAME	TYPE	DESCRIPTION
maxLoops	int	Maximum number of tool loops. Must be at least 1.

Return

( boolean )

True if the value was applied successfully, false if the value is invalid.

mcp

mcp(configs: Values) : void

Description

Configures the MCP (Model Context Protocol) servers to use in chat and stream operations. Each server exposes tools that the model can invoke automatically during the conversation. Tools are available with the prefix serverName__toolName.

Supported transport types:

remote: connects to an MCP server via HTTP Streamable (SSE/HTTP)
stdio: starts a local process and communicates via stdin/stdout

How To Use

// Remote MCP server via HTTP
const servers = _val.list()
    .add(
        _val.map()
            .set('type', 'remote')
            .set('name', 'myServer')
            .set('url', 'https://mcp.example.com')
            .set('endpoint', '/mcp')
            .set('headers',
                _val.map().set('Authorization', 'Bearer YOUR_TOKEN')
            )
    )

client.mcp(servers)

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Use the available tool.'))

const response = client.chat(messages)
_out.json(response.toJSON())

Attributes

NAME	TYPE	DESCRIPTION
configs	Values	List of MCP server configurations. Each entry is an object with the following fields:

		Common fields:
		- `type` (required): transport type — `remote` or `stdio`
		- `name` (optional): server name, used as a prefix for tools. If omitted, it is auto-generated

		For `type: remote`:
		- `url` (required): base URL of the MCP server, e.g. `https://mcp.example.com`
		- `endpoint` (optional): MCP endpoint path. Default: `/mcp`
		- `headers` (optional): object with additional HTTP headers, e.g. `Authorization`

		For `type: stdio`:
		- `command` (required): command to execute
		- `args` (optional): list of command arguments
		- `env` (optional): object with environment variables

Return

( void )

model

model(model: string) : boolean

Description

Sets the AI model to use in chat, stream and embeddings operations. The model is validated against the list of available models on the provider.

How To Use

const ok = client.model('gpt-4o')
if (!ok) {
    _log.error('Invalid or unavailable model.')
}

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use, for example: `gpt-4o`.

Return

( boolean )

True if the model is valid and was set, false otherwise.

models

models() : Values

Description

Lists all models available on the configured AI provider.

How To Use

const models = client.models()
_out.json(modelos.toJSON())

Return

( Values )

List of available models, each as an object with its metadata.

provider

provider(provider: string) : boolean

Description

Switches the AI provider and reinitializes the client with the new provider settings defined in the application configuration file.

How To Use

const switched = client.provider('anthropic')
if (switched) {
    _log.info('Provider switched successfully.')
}

Attributes

NAME	TYPE	DESCRIPTION
provider	string	Name of the AI provider as defined in the application settings.

Return

( boolean )

True if the provider was switched successfully, false otherwise.

stream

stream(model: string, messages: Values, onToken: java.util.function.Consumer<Values>) : void

Description

Runs a streaming conversation explicitly specifying the model to use, overriding the default configured model, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Tell me a short story.'))

client.stream('gpt-4o-mini', messages, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
})

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.

Return

( void )

stream(model: string, messages: Values, onToken: java.util.function.Consumer<Values>, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : void

Description

Runs a streaming conversation explicitly specifying the model to use, with MCP tool support via callback, overriding the default configured model, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'What time is it?'))

client.stream('gpt-4o-mini', messages, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
}, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( void )

stream(model: string, messages: Values, options: Values, onToken: java.util.function.Consumer<Values>) : void

Description

Runs a streaming conversation explicitly specifying the model to use, with additional options, overriding the default configured model, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

client.stream('gpt-4o-mini', messages, options, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
})

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.

Return

( void )

stream(model: string, messages: Values, options: Values, onToken: java.util.function.Consumer<Values>, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : void

Description

Runs a streaming conversation explicitly specifying the model to use, with additional options and MCP tool support via callback, overriding the default configured model, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

client.stream('gpt-4o-mini', messages, options, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
}, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})

Attributes

NAME	TYPE	DESCRIPTION
model	string	Identifier of the model to use in this call.
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( void )

stream(messages: Values, onToken: java.util.function.Consumer<Values>) : void

Description

Runs a streaming conversation with the configured AI model, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Tell me a short story.'))

client.stream(messages, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
})

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.

Return

( void )

stream(messages: Values, onToken: java.util.function.Consumer<Values>, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : void

Description

Runs a streaming conversation with the configured AI model, with MCP tool support via callback, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'What time is it?'))

client.stream(messages, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
}, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( void )

stream(messages: Values, options: Values, onToken: java.util.function.Consumer<Values>) : void

Description

Runs a streaming conversation with the configured AI model, with additional options, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

client.stream(messages, options, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
})

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.

Return

( void )

stream(messages: Values, options: Values, onToken: java.util.function.Consumer<Values>, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : void

Description

Runs a streaming conversation with the configured AI model, with additional options and MCP tool support via callback, processing each token as it is generated.

How To Use

const messages = _val.list()
    .add(_val.map().set('role', 'user').set('content', 'Hello!'))

const options = _val.map().set('temperature', 0.7)

client.stream(messages, options, (chunk) => {
    _out.print(chunk.get('choices').get(0).get('delta').get('content'))
}, (toolName, args, mcpClient, tool) => {
    _log.info('Tool invoked: ' + toolName)
    return null
})

Attributes

NAME	TYPE	DESCRIPTION
messages	Values	List of conversation messages.
options	Values	Additional options: `temperature` (0.0–2.0), `max_tokens`, `top_p`.
onToken	java.util.function.Consumer	Callback invoked for each token received, receiving the response chunk as argument.
toolCallback	org.netuno.tritao.ai.client.Client$ToolCallback	Callback invoked before each tool execution. Return null for normal execution or a Values to override the result.

Return

( void )

chat​

chat(model: string, messages: Values) : Values​

Description​

How To Use​

Attributes​

Return​

chat(model: string, messages: Values, options: Values) : Values​

Description​

How To Use​

Attributes​

Return​

chat(model: string, messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values​

Description​

How To Use​

Attributes​

Return​

chat(model: string, messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values​

Description​

How To Use​

Attributes​

Return​

chat(messages: Values) : Values​

Description​

How To Use​

Attributes​

Return​

chat(messages: Values, options: Values) : Values​

Description​

How To Use​

Attributes​

Return​

chat(messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values​

Description​

How To Use​

Attributes​

Return​

chat(messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings​

embeddings(input: string) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(model: string, input: string) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(model: string, input: string, options: Values) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(model: string, inputs: Values) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(model: string, inputs: Values, options: Values) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(inputs: Values) : Values​

Description​

How To Use​

Attributes​

Return​

embeddings(inputs: Values, options: Values) : Values​

Description​

How To Use​

Attributes​

Return​

getMaxToolLoops​

getMaxToolLoops() : int​

Description​

chat

chat(model: string, messages: Values) : Values

Description

How To Use

Attributes

Return

chat(model: string, messages: Values, options: Values) : Values

Description

How To Use

Attributes

Return

chat(model: string, messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

How To Use

Attributes

Return

chat(model: string, messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

How To Use

Attributes

Return

chat(messages: Values) : Values

Description

How To Use

Attributes

Return

chat(messages: Values, options: Values) : Values

Description

How To Use

Attributes

Return

chat(messages: Values, options: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

How To Use

Attributes

Return

chat(messages: Values, toolCallback: org.netuno.tritao.ai.client.Client$ToolCallback) : Values

Description

How To Use

Attributes

Return

embeddings

embeddings(input: string) : Values

Description

How To Use

Attributes

Return

embeddings(model: string, input: string) : Values

Description

How To Use

Attributes

Return

embeddings(model: string, input: string, options: Values) : Values

Description

How To Use

Attributes

Return

embeddings(model: string, inputs: Values) : Values

Description

How To Use

Attributes

Return

embeddings(model: string, inputs: Values, options: Values) : Values

Description

How To Use

Attributes

Return

embeddings(inputs: Values) : Values

Description

How To Use

Attributes

Return

embeddings(inputs: Values, options: Values) : Values

Description

How To Use

Attributes

Return

getMaxToolLoops

getMaxToolLoops() : int

Description