/openai-fsharp

OpenAI - F# - A library that allows you to interact with OpenAI API in an easy, fluent, and typed way!

Primary LanguageF#MIT LicenseMIT

OpenAI F#

GitHub Workflow Status (main) License NuGet Version NuGet Version

OpenAI F# is a library that allows you to interact with Open AI API in an easy, fluent, and typed way!

Get Started

Add the library to your project using DotNet CLI:

dotnet add package OpenAI.Client

Or using Paket:

paket add OpenAI.Client

The library provide two styles to consume the APIs, either a methods calls through pipe operator as shown below:

open OpenAI
open OpenAI.Client
open OpenAI.Models

let client =
    Config(
        { Endpoint = "https://api.openai.com/v1"
          ApiKey = "your-api-key" },
        HttpRequester()
    )
    
let result = client |> models |> list

Or using a computation expression builder object as shown below:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        models
        list
    }

You can check OpenAI API reference at API reference.

Usage

Models Resource

List and describe the various models available in the API. You can refer to the Models documentation to understand what models are available and the differences between them.

list

Lists the currently available models, and provides basic information about each one such as the owner and availability.

let result = client |> models |> list

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        models
        list
    }

retrieve

Retrieves a model instance, providing basic information about the model such as the owner and permissioning.

let result = client |> models |> retrieve "text-davinci-003"

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        models
        retrieve "text-davinci-003"
    }

delete

Delete a fine-tuned model.

let result = client |> models |> delete "curie:ft-acmeco-2021-03-03-21-44-20"

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        models
        delete "curie:ft-acmeco-2021-03-03-21-44-20"
    }

Completions Resource

Given a prompt, the model will return one or more predicted completions, and can also return the probabilities of alternative tokens at each position.

create

Creates a completion for the provided prompt and parameters.

let result =
    client
    |> completions
    |> Completions.create
        { Model = "text-davinci-003"
          Prompt = "What is the meaning of living?"
          Temperature = 0.5
          Stop = "." }

Or using a computation expression builder:

let result =
    openAI {
     endPoint "https://api.openai.com/v1"
     apiKey "your-api-key"
     completions
     create
        { Model = "text-davinci-003"
          Prompt = "What is the meaning of living?"
          Temperature = 0.5
          Stop = "." }
    }

Chat Resource

Given a chat conversation, the model will return a chat completion response.

create

Creates a completion for the chat message.

let result =
    client
    |> chat
    |> create
      { Model = "text-davinci-003"
        Messages = [| {Role = "user"; Content = "Hello!"} |] }

Or using a computation expression builder:

let result =
    openAI {
    endPoint (url "")
    apiKey "apiKey"
    chat
    create
      { Model = "text-davinci-003"
        Messages = [| {Role = "user"; Content = "Hello!"} |] }
    }

Edits Resource

Given a prompt and an instruction, the model will return an edited version of the prompt.

create

Creates a new edit for the provided input, instruction, and parameters.

let result =
    client
    |> edits
    |> Edits.create
           { Model = "text-davinci-edit-001"
             Input = "What day of the wek is it?"
             Instruction = "Fix the spelling mistakes" }

Or using a computation expression builder:

let result =
    openAI {
     endPoint "https://api.openai.com/v1"
     apiKey "your-api-key"
     edits
     create
        { Model = "text-davinci-edit-001"
         Input = "What day of the wek is it?"
         Instruction = "Fix the spelling mistakes" }
    }

Images Resource

Given a prompt and/or an input image, the model will generate a new image.

create

Creates an image given a prompt.

let result =
    client
    |> images
    |> create
         { Prompt = "A cute baby sea otter"
           N = 2
           Size = "1024x1024"
           responseFormat = "url" }

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        images
        create
            { Prompt = "A cute baby sea otter"
              N = 2
              Size = "1024x1024"
              responseFormat = "url" }
    }

edit

Creates an edited or extended image given an original image and a prompt.

let result =
    client
    |> images
    |> edit
        { Image = "sample.png"
          Mask = "sample.png"
          Prompt = "A cute baby sea otter wearing a beret"
          N = 2
          Size = "1024x1024"
          responseFormat = "url" }

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        images
        edit
            { Image = "sample.png"
              Mask = "sample.png"
              Prompt = "A cute baby sea otter wearing a beret"
              N = 2
              Size = "1024x1024"
              responseFormat = "url" }
    }

variation

Creates a variation of a given image.

let result =
    client
    |> images
    |> variation
        { Image = "sample.png"
          N = 2
          Size = "1024x1024"
          responseFormat = "url" }

Or using a computation expression builder:

let result =
    openAI {
     endPoint "https://api.openai.com/v1"
     apiKey "your-api-key"
     images
     variation
        { Image = "sample.png"
          N = 2
          Size = "1024x1024"
          responseFormat = "url" }
    }

Embeddings Resource

Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms.

create

Creates an embedding vector representing the input text.

let result =
    client
    |> embeddings
    |> create
        { Model = "text-embedding-ada-002"
          Input = "The food was delicious and the waiter..." }

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        embeddings
        create
             { Model = "text-embedding-ada-002"
               Input = "The food was delicious and the waiter..." }
    }

Audio Resource

Turn audio into text.

transcript

Transcribes audio into the input language.

let result =
    client
    |> audio
    |> transcript
      { TranscriptRequest.Default with
          File = "sample-audio.mp3"
          Model = "whisper-1" }

Or using a computation expression builder:

let result =
    openAI {
      endPoint (url "")
      apiKey "apiKey"
      audio
      transcript
          { TranscriptRequest.Default with
              File = "sample-audio.mp3"
              Model = "whisper-1" }
    }

translate

Translates audio into into English.

let result =
    client
    |> audio
    |> translate
      { File = "sample-audio.mp3"
        Model = "whisper-1" }

Or using a computation expression builder:

let result =
    openAI {
      endPoint (url "")
      apiKey "apiKey"
      audio
      translate
          { File = "sample-audio.mp3"
            Model = "whisper-1" }
    }

Files Resource

Files are used to upload documents that can be used with features like Fine-tuning.

list

Returns a list of files that belong to the user's organization.

let result = client |> files |> list

Or using a computation expression builder:

let result =
     openAI {
         endPoint "https://api.openai.com/v1"
         apiKey "your-api-key"
         files
         list
     }

upload

Upload a file that contains document(s) to be used across various endpoints/features.

let result =
    client
    |> files
    |> upload
        { File = @"sample-json.txt"
          Purpose = "fine-tune" }

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        files
        upload
             { File = @"sample-json.txt"
               Purpose = "fine-tune" }
    }

delete

Delete a file.

let result =
    client
    |> files
    |> delete "file-qtUwySute1Zf2yT6mWIGTCwq"

Or using a computation expression builder:

let result =
    openAI {
         endPoint "https://api.openai.com/v1"
         apiKey "your-api-key"
         files
         delete "file-qtUwySute1Zf2yT6mWIGTCwq"
    }

retrieve

Returns information about a specific file.

let result = client |> files |> retrieve "file-Lpe0n5tOHtoG6OVVbk5d4iXA"

Or using a computation expression builder:

let result =
    openAI {
         endPoint "https://api.openai.com/v1"
         apiKey "your-api-key"
         files
         retrieve "file-Lpe0n5tOHtoG6OVVbk5d4iXA"
    }

download

Returns the contents of the specified file.

let result = client |> files |> download "file-Lpe0n5tOHtoG6OVVbk5d4iXA"

Or using a computation expression builder:

let result =
    openAI {
         endPoint "https://api.openai.com/v1"
         apiKey "your-api-key"
         files
         download "file-Lpe0n5tOHtoG6OVVbk5d4iXA"
    }

FineTunes Resource

Manage fine-tuning jobs to tailor a model to your specific training data.

create

Creates a job that fine-tunes a specified model from a given dataset.

let result =
    client
    |> fineTunes
    |> create { CreateRequest.Default with TrainingFile = "file-Lpe0n5tOHtoG6OVVbk5d4iXA" }

Or using a computation expression builder:

let result = 
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        fineTunes
        create { CreateRequest.Default with TrainingFile = "file-Lpe0n5tOHtoG6OVVbk5d4iXA" }
    }

list

List your organization's fine-tuning jobs.

let result = client |> fineTunes |> list

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        fineTunes
        list
    }

retrieve

Gets info about the fine-tune job.

let result = client |> fineTunes |> retrieve "ft-kNJeZjAhlmFLdCuIzcpkrmNI"

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        retrieve
        retrieve "ft-kNJeZjAhlmFLdCuIzcpkrmNI"
    }

cancel

Immediately cancel a fine-tune job.

let result = client |> fineTunes |> cancel "ft-kNJeZjAhlmFLdCuIzcpkrmNI"

Or using a computation expression builder:

let result =
 openAI {
     endPoint "https://api.openai.com/v1"
     apiKey "your-api-key"
     fineTunes
     cancel "ft-kNJeZjAhlmFLdCuIzcpkrmNI"
 }

list events

Get fine-grained status updates for a fine-tune job.

let result = client |> fineTunes |> listEvents "ft-kNJeZjAhlmFLdCuIzcpkrmNI"

Or using a computation expression builder:

let result =
    openAI {
        endPoint "https://api.openai.com/v1"
        apiKey "your-api-key"
        fineTunes
        listEvents "ft-kNJeZjAhlmFLdCuIzcpkrmNI"
    }

Moderations Resource

Given a input text, outputs if the model classifies it as violating OpenAI's content policy.

create

Classifies if text violates OpenAI's Content Policy.

let result =
 client
 |> moderations
 |> create
     { Model = "text-moderation-latest"
       Input = "I want to kill them." }

Or using a computation expression builder:

let result =
 openAI {
     endPoint "https://api.openai.com/v1"
     apiKey "your-api-key"
     moderations
     create
         { Model = "text-moderation-latest"
           Input = "I want to kill them." }
 }

Credits

This library was inspired from openai-php and FsHttp.