OpenAI API client for Kotlin

Kotlin client for OpenAI's API with multiplatform and coroutines capabilities.

🛠 Setup

1. Install OpenAI API Kotlin client by adding the following dependency to your gradle.build file:

repositories {
    mavenCentral()
}

dependencies {
    implementation "com.aallam.openai:openai-client:<version>"
}

2. Choose and add to your dependencies one of Ktor's engines.

Alternatively, you can use openai-client-bom

Multiplaform

In multiplatform projects, add openai client dependency to commonMain, and choose an engine for each target.

💡 Getting Started

Create an instance of OpenAI client:

val openAI = OpenAI(apiKey)

Use your OpenAI instance to make API requests:

List models

val models: List<Model> = openAI.models()

Retrieve an model

val id = ModelId("text-ada-001")
val model: Model = openAI.model(id)

Create completion

val completionRequest = CompletionRequest(
    model = ModelId("text-ada-001"),
    prompt = "Somebody once told me the world is gonna roll me",
    echo = true
)
val completion: TextCompletion = openAI.completion(Ada, completionRequest)

Create completion stream

val request = CompletionRequest(
    model = ModelId("text-davinci-002"),
    prompt = "Once upon a time",
    maxTokens = 5,
    temperature = 1.0,
    topP = 1.0,
    n = 1,
    stop = listOf("\n"),
)
val completions: Flow<TextCompletion> = openAI.completions(request)

Create edits

val edit = openAI.edit(
    request = EditsRequest(
        model = ModelId("text-davinci-edit-001"),
        input = "What day of the wek is it?",
        instruction = "Fix the spelling mistakes"
    )
)

List files

val files: List<File> = openAI.files()

Create embeddings

val embeddings: List<Embedding> = openAI.embeddings(
    request = EmbeddingRequest(
        model = ModelId("text-similarity-babbage-001"),
        input = listOf("The food was delicious and the waiter...")
    )
)

Create moderation

val moderation = openAI.moderations(
    request = ModerationRequest(
        input = "I want to kill them."
    )
)

Create fine-tunes

val fineTune = openAI.fineTune(
    request = FineTuneRequest(
        trainingFile = trainingFile,
        model = ModelId("ada")
    )
)

Create images

val images = openAI.image(
    creation = ImageCreationURL(
        prompt = "A cute baby sea otter",
        n = 2,
        size = ImageSize.is1024x1024
    )
)

Edit images

val images = openAI.image(
    edit = ImageEditURL( // or 'ImageEditJSON'
        image = FilePath(imagePath),
        mask = FilePath(maskPath),
        prompt = "a sunlit indoor lounge area with a pool containing a flamingo",
        n = 1,
        size = ImageSize.is1024x1024
    )
)

Create image variants

val images = openAI.image(
    variation = ImageVariationURL( // or, 'ImageVariationJSON'
        image = FilePath(imagePath),
        n = 1,
        size = ImageSize.is1024x1024
    )
)

ℹ️ Sample apps

Sample apps are available under sample, please check the README for running instructions.

📄 License

OpenAI Kotlin API Client is an open-sourced software licensed under the MIT license. This is an unofficial library, it is not affiliated with nor endorsed by OpenAI. Contributions are welcome.

⭐️ Support