Update Pipe langbase.pipes.update()

interface PipeUpdateOptions {
	name: string;
	description?: string;
	status?: 'public' | 'private';
	model?: string;
	stream?: boolean;
	json?: boolean;
	store?: boolean;
	moderate?: boolean;
	top_p?: number;
	max_tokens?: number;
	temperature?: number;
	presence_penalty?: number;
	frequency_penalty?: number;
	stop?: string[];
	tools?: {
		type: 'function';
		function: {
			name: string;
			description?: string;
			parameters?: Record<string, any>;
		};
	}[];
	tool_choice?: 'auto' | 'required' | ToolChoice;
	parallel_tool_calls?: boolean;
	messages?: Message[];
	variables?: Record<string, string>;
	memory?: {
		name: string;
	}[];
	response_format?: ResponseFormat;
}

Following are the properties of the options object.

Name of the pipe to update.

Description of the AI pipe.

Status of the pipe. Defaults to public

Pipe LLM model. Combination of model provider and model id.

Format: provider:model_id

You can copy the ID of a model from the list of supported LLM models at Langbase.

Default: openai:gpt-4o-mini

Tool usage configuration.

An array of memories that the Pipe should use

interface Memory {
	name: string;
}

If enabled, the output will be streamed in real-time like ChatGPT. This is helpful if user is directly reading the text.

Default: true

Enforce the output to be in JSON format.

Default: false

If enabled, both prompt and completions will be stored in the database. Otherwise, only system prompt and few shot messages will be saved.

Default: true

If enabled, Langbase blocks flagged requests automatically.

Default: false

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random. Lower values like 0.2 will make it more focused and deterministic.

Default: 0.7

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

Default: 1

Maximum number of tokens in the response message returned.

Default: 1000

Penalizes a word based on its occurrence in the input text.

Default: 1

Penalizes a word based on how frequently it appears in the training data.

Default: 1

Up to 4 sequences where the API will stop generating further tokens.

Default: []

Call multiple tools in parallel, allowing the effects and results of these function calls to be resolved in parallel.

Default: true

A messages array including the following properties. Optional if variables are provided.

interface Message {
	role: 'user' | 'assistant' | 'system'| 'tool';
	content: string | null;
	name?: string;
	tool_call_id?: string;
	tool_calls?: ToolCall[];
}

An object containing pipe variables. The key is the variable name and the value is the variable value.

Default: {}

Defines the format of the response. Primarily used for Structured Outputs. To enforce Structured Outputs, set type to json_schema, and provide a JSON schema for your response with strict: true option.

Default: text

 type ResponseFormat =
	| {type: 'text'}
	| {type: 'json_object'}
	| {
		type: 'json_schema';
		json_schema: {
			description?: string;
			name: string;
			schema?: Record<string, unknown>;
			strict?: boolean | null;
		};
	};

The response object returned by the langbase.pipes.update() function.

interface PipeUpdateResponse {
	name: string;
	description: string;
	status: 'public' | 'private';
	owner_login: string;
	url: string;
	type: 'chat' | 'generate' | 'run';
	api_key: string;
}