Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/henrymxu/openai-kotlin
Unofficial OpenAi API in Kotlin Multiplatform (Android, iOS, JVM, JS)
https://github.com/henrymxu/openai-kotlin
kotlin kotlin-multiplatform openai-api
Last synced: about 2 months ago
JSON representation
Unofficial OpenAi API in Kotlin Multiplatform (Android, iOS, JVM, JS)
- Host: GitHub
- URL: https://github.com/henrymxu/openai-kotlin
- Owner: henrymxu
- License: mit
- Created: 2022-12-04T07:40:23.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2022-12-14T01:55:24.000Z (about 2 years ago)
- Last Synced: 2023-08-09T09:43:04.224Z (over 1 year ago)
- Topics: kotlin, kotlin-multiplatform, openai-api
- Language: Kotlin
- Homepage:
- Size: 1.75 MB
- Stars: 2
- Watchers: 1
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# OpenAI Kotlin Multiplatform Library
The Unofficial OpenAI Kotlin library provides convenient access to the OpenAI API for multiple languages
via [Kotlin Multiplatform Module](https://kotlinlang.org/docs/multiplatform.html).
This library contains support for the following platforms:
- Android
- iOS
- JVM
- JS
It includes statically defined types for both requests and responses for all the publicly documented APIs.
## Usage
Using the library requires an OpenAI API Key!
If you do not have an api key already, follow these steps:
1. Create an account at https://beta.openai.com/
2. Navigate to the API Keys [page](https://beta.openai.com/account/api-keys)
3. Create a new secret key
The following is a simple example to fetch the details of a model
```kotlin
val client = OpenAiClient.Builder(apiKey = "") {
// optional configurations
}
val model = client.api.retrieveModel("text-davinci-003")
assertEquals("text-davinci-003", model.id)
```
Next is a more complex example of streaming the results of a completion request
```kotlin
val model = "text-davinci-003"
val request = CreateCompletionRequest(
model,
prompt = "Say this is a test",
logprobs = null,
stop = listOf("\n")
)
val result = client.api.streamCreateCompletion(request)
CoroutineScope(Dispatchers.IO).launch {
result.collect { completion ->
// ... do stuff here
}
}
```
## Examples
## Android
### Setup
1. Navigate to the `android` module
2. Make a copy of the `.env.template` file and name it `.env`
3. Populate the required values for each key in `.env`
- NOTE: Do **NOT** rename any keys!
- NOTE: Do **NOT** commit the `.env` file!
4. Run application
## iOS
### Setup
1. Navigate to the `ios` module
2. Make a copy of the `Config.xcconfig.template` file and name it `Config.xcconfig`
3. Populate the required values for each key in `Config.xcconfig`
- NOTE: Do **NOT** rename any keys!
- NOTE: Do **NOT** commit the `Config.xcconfig` file!
4. Run application
TBA
## NodeJS
### Setup
1. Navigate to the `web` module
2. Make a copy of the `.env.template` file and name it `.env`
3. Populate the required values for each key in `.env`
- NOTE: Do **NOT** rename any keys (they must be prefixed with `REACT_APP`)!
- NOTE: Do **NOT** commit the `.env` file!
4. Execute `npm run start`
To access variables of type `Long` (e.g `created` / `createdAt`), you must access the `a1_1`
of the variable.
```javascript
data.created.a1_1
```
This is simply a result of `kotlin` not being able to export the `Long` type to `JS` yet.
## Tests
# References
- [OpenAI API Spec](https://github.com/openai/openai-openapi)