Class: PromptTemplate

prompts.PromptTemplate

Schema to represent a basic prompt for an LLM.

Example

import { PromptTemplate } from "langchain/prompts";

const prompt = new PromptTemplate({
  inputVariables: ["foo"],
  template: "Say {foo}",
});

Hierarchy

• BaseStringPromptTemplate

  ↳ PromptTemplate

Implements

• PromptTemplateInput

Constructors

constructor

• new PromptTemplate(input)

Parameters

Name	Type
input	PromptTemplateInput

Overrides

BaseStringPromptTemplate.constructor

Defined in

langchain/src/prompts/prompt.ts:61

Properties

inputVariables

• inputVariables: string[]

A list of variable names the prompt template expects

Implementation of

PromptTemplateInput.inputVariables

Inherited from

BaseStringPromptTemplate.inputVariables

Defined in

langchain/src/prompts/base.ts:51

---

outputParser

• Optional outputParser: BaseOutputParser

How to parse the output of calling an LLM on this formatted prompt

Implementation of

PromptTemplateInput.outputParser

Inherited from

BaseStringPromptTemplate.outputParser

Defined in

langchain/src/prompts/base.ts:53

---

partialVariables

• Optional partialVariables: InputValues

Partial variables

Implementation of

PromptTemplateInput.partialVariables

Inherited from

BaseStringPromptTemplate.partialVariables

Defined in

langchain/src/prompts/base.ts:55

---

template

• template: string

The propmt template

Implementation of

PromptTemplateInput.template

Defined in

langchain/src/prompts/prompt.ts:55

---

templateFormat

• templateFormat: TemplateFormat = "f-string"

The format of the prompt template. Options are 'f-string', 'jinja-2'

Default Value

'f-string'

Implementation of

PromptTemplateInput.templateFormat

Defined in

langchain/src/prompts/prompt.ts:57

---

validateTemplate

• validateTemplate: boolean = true

Whether or not to try validating the template on initialization

Default Value

true

Implementation of

PromptTemplateInput.validateTemplate

Defined in

langchain/src/prompts/prompt.ts:59

Methods

_getPromptType

▸ _getPromptType(): "prompt"

Return the string type key uniquely identifying this class of prompt template.

Returns

"prompt"

Overrides

BaseStringPromptTemplate._getPromptType

Defined in

langchain/src/prompts/prompt.ts:80

---

format

▸ format(values): Promise<string>

Format the prompt given the input values.

Example

prompt.format({ foo: "bar" });

Parameters

Name	Type	Description
values	InputValues	A dictionary of arguments to be passed to the prompt template.

Returns

Promise<string>

A formatted prompt string.

Overrides

BaseStringPromptTemplate.format

Defined in

langchain/src/prompts/prompt.ts:84

---

formatPromptValue

▸ formatPromptValue(values): Promise<BasePromptValue>

Format the prompt given the input values and return a formatted prompt value.

Parameters

Name	Type
values	InputValues

Returns

Promise<BasePromptValue>

A formatted PromptValue.

Inherited from

BaseStringPromptTemplate.formatPromptValue

Defined in

langchain/src/prompts/base.ts:152

---

mergePartialAndUserVariables

▸ mergePartialAndUserVariables(userVariables): Promise<InputValues>

Parameters

Name	Type
userVariables	InputValues

Returns

Promise<InputValues>

Inherited from

BaseStringPromptTemplate.mergePartialAndUserVariables

Defined in

langchain/src/prompts/base.ts:69

---

partial

▸ partial(values): Promise<PromptTemplate>

Parameters

Name	Type
values	PartialValues

Returns

Promise<PromptTemplate>

Overrides

BaseStringPromptTemplate.partial

Defined in

langchain/src/prompts/prompt.ts:141

---

serialize

▸ serialize(): SerializedPromptTemplate

Return a json-like object representing this prompt template.

Returns

SerializedPromptTemplate

Overrides

BaseStringPromptTemplate.serialize

Defined in

langchain/src/prompts/prompt.ts:153

---

deserialize

▸ Static deserialize(data): Promise<PromptTemplate>

Load a prompt template from a json-like object describing it.

Remarks

Deserializing needs to be async because templates (e.g. FewShotPromptTemplate) can reference remote resources that we read asynchronously with a web request.

Parameters

Name	Type
data	SerializedPromptTemplate

Returns

Promise<PromptTemplate>

Overrides

BaseStringPromptTemplate.deserialize

Defined in

langchain/src/prompts/prompt.ts:167

---

fromExamples

▸ Static fromExamples(examples, suffix, inputVariables, exampleSeparator?, prefix?): PromptTemplate

Take examples in list format with prefix and suffix to create a prompt.

Intendend to be used a a way to dynamically create a prompt from examples.

Parameters

Name	Type	Default value	Description
examples	string[]	undefined	List of examples to use in the prompt.
suffix	string	undefined	String to go after the list of examples. Should generally set up the user's input.
inputVariables	string[]	undefined	A list of variable names the final prompt template will expect
exampleSeparator	string	"\n\n"	The separator to use in between examples
prefix	string	""	String that should go before any examples. Generally includes examples.

Returns

PromptTemplate

The final prompt template generated.

Defined in

langchain/src/prompts/prompt.ts:102

---

fromTemplate

▸ Static fromTemplate(template, «destructured»?): PromptTemplate

Load prompt template from a template f-string

Parameters

Name	Type
template	string
«destructured»	Omit<PromptTemplateInput, "template" | "inputVariables">

Returns

PromptTemplate

Defined in

langchain/src/prompts/prompt.ts:119