Package org.springframework.ai.openai.api

Class OpenAiApi

java.lang.Object

org.springframework.ai.openai.api.OpenAiApi

public class OpenAiApi
extends Object

Single class implementation of the OpenAI Chat Completion API: https://platform.openai.com/docs/api-reference/chat and OpenAI Embedding API: https://platform.openai.com/docs/api-reference/embeddings.

Author:

Christian Tzolov, Michael Lavelle

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final record	OpenAiApi.ChatCompletion	Represents a chat completion response returned by model, based on the provided input.
static final record	OpenAiApi.ChatCompletionChunk	Represents a streamed chunk of a chat completion response returned by model, based on the provided input.
static enum	OpenAiApi.ChatCompletionFinishReason	The reason the model stopped generating tokens.
static final record	OpenAiApi.ChatCompletionMessage	Message comprising the conversation.
static final record	OpenAiApi.ChatCompletionRequest	Creates a model response for the given chat conversation.
static enum	OpenAiApi.ChatModel	OpenAI Chat Completion Models: GPT-4 and GPT-4 Turbo and GPT-3.5 Turbo.
static final record	OpenAiApi.Embedding	Represents an embedding vector returned by embedding endpoint.
static final record	OpenAiApi.EmbeddingList<T>	List of multiple embedding responses.
static enum	OpenAiApi.EmbeddingModel	OpenAI Embeddings Models: Embeddings.
static final record	OpenAiApi.EmbeddingRequest<T>	Creates an embedding vector representing the input text.
static final record	OpenAiApi.FunctionTool	Represents a tool the model may call.
static final record	OpenAiApi.LogProbs	Log probability information for the choice.
static final record	OpenAiApi.Usage	Usage statistics for the completion request.

Field Summary

Fields

Modifier and Type	Field	Description
static final String	DEFAULT_CHAT_MODEL
static final String	DEFAULT_EMBEDDING_MODEL

Constructor Summary

Constructors

Constructor	Description
OpenAiApi(String openAiToken)	Create an new chat completion api with base URL set to https://api.openai.com
OpenAiApi(String baseUrl, String openAiToken)	Create a new chat completion api.
OpenAiApi(String baseUrl, String openAiToken, org.springframework.web.client.RestClient.Builder restClientBuilder)	Create a new chat completion api.
OpenAiApi(String baseUrl, String openAiToken, org.springframework.web.client.RestClient.Builder restClientBuilder, org.springframework.web.client.ResponseErrorHandler responseErrorHandler)	Create a new chat completion api.

Method Summary

Modifier and Type	Method	Description
org.springframework.http.ResponseEntity<OpenAiApi.ChatCompletion>	chatCompletionEntity(OpenAiApi.ChatCompletionRequest chatRequest)	Creates a model response for the given chat conversation.
reactor.core.publisher.Flux<OpenAiApi.ChatCompletionChunk>	chatCompletionStream(OpenAiApi.ChatCompletionRequest chatRequest)	Creates a streaming chat response for the given chat conversation.
<T> org.springframework.http.ResponseEntity<OpenAiApi.EmbeddingList<OpenAiApi.Embedding>>	embeddings(OpenAiApi.EmbeddingRequest<T> embeddingRequest)	Creates an embedding vector representing the input text or token array.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

DEFAULT_CHAT_MODEL

public static final String DEFAULT_CHAT_MODEL

DEFAULT_EMBEDDING_MODEL

public static final String DEFAULT_EMBEDDING_MODEL

Constructor Details

OpenAiApi

public OpenAiApi(String openAiToken)

Create an new chat completion api with base URL set to https://api.openai.com

Parameters:

openAiToken - OpenAI apiKey.

OpenAiApi

public OpenAiApi(String baseUrl,
 String openAiToken)

Create a new chat completion api.

Parameters:

baseUrl - api base URL.

openAiToken - OpenAI apiKey.

OpenAiApi

public OpenAiApi(String baseUrl,
 String openAiToken,
 org.springframework.web.client.RestClient.Builder restClientBuilder)

Create a new chat completion api.

Parameters:

baseUrl - api base URL.

openAiToken - OpenAI apiKey.

restClientBuilder - RestClient builder.

OpenAiApi

public OpenAiApi(String baseUrl,
 String openAiToken,
 org.springframework.web.client.RestClient.Builder restClientBuilder,
 org.springframework.web.client.ResponseErrorHandler responseErrorHandler)

Create a new chat completion api.

Parameters:

baseUrl - api base URL.

openAiToken - OpenAI apiKey.

restClientBuilder - RestClient builder.

responseErrorHandler - Response error handler.

Method Details

chatCompletionEntity

public org.springframework.http.ResponseEntity<OpenAiApi.ChatCompletion> chatCompletionEntity(OpenAiApi.ChatCompletionRequest chatRequest)

Creates a model response for the given chat conversation.

Parameters:

chatRequest - The chat completion request.

Returns:

Entity response with OpenAiApi.ChatCompletion as a body and HTTP status code and headers.

chatCompletionStream

public reactor.core.publisher.Flux<OpenAiApi.ChatCompletionChunk> chatCompletionStream(OpenAiApi.ChatCompletionRequest chatRequest)

Creates a streaming chat response for the given chat conversation.

Parameters:

chatRequest - The chat completion request. Must have the stream property set to true.

Returns:

Returns a Flux stream from chat completion chunks.

embeddings

public <T>
org.springframework.http.ResponseEntity<OpenAiApi.EmbeddingList<OpenAiApi.Embedding>> embeddings(OpenAiApi.EmbeddingRequest<T> embeddingRequest)

Creates an embedding vector representing the input text or token array.

Type Parameters:

T - Type of the entity in the data list. Can be a String or List of tokens (e.g. Integers). For embedding multiple inputs in a single request, You can pass a List of String or List of List of tokens. For example:

List.of("text1", "text2", "text3") or List.of(List.of(1, 2, 3), List.of(3, 4, 5)) 

Parameters:

embeddingRequest - The embedding request.

Returns:

Returns list of OpenAiApi.Embedding wrapped in OpenAiApi.EmbeddingList.