The fal.ai Kotlin Client

About the Project

The FalClient is a robust and user-friendly Kotlin package designed for seamless integration of fal serverless functions into Kotlin projects, including Android Platform.

Getting Started

The FalClient library serves as a client for fal serverless Python functions. Before using this library, ensure you've got your fal key from our dashboard.

Install

Add falclient dependency to your build file. Here is the build.gradle.kts example:

implementation("ai.fal:falclient:0.1.1")

Client Library

import ai.fal.falclient.*

// Initialize FalClient with your credentials
val authKey = "$FAL_KEY"
val falClient = FalClient(ClientConfig(credentials = authKey))

// Define the input parameters for the function
val inputParams = hashMapOf(
    "prompt" to "a cute shih-tzu puppy",
    "model_name" to "stabilityai/stable-diffusion-xl-base-1.0",
    "image_size" to "square_hd"
)

// Run the function and print the result
try {
    val result = falClient.run("110602490-lora", inputParams)
    println(result)
} catch (e: Error) {
    println("Error running the function: ${e.message}")
}

// Subscribe to the function and print the updates
try {
    val subscribeResult = falClient.subscribe("110602490-lora", inputParams, 1000L, 30000L, true) { update ->
        println("Update: $update")
    }
    println("Final result: $subscribeResult")
} catch (e: Error) {
    println("Error subscribing to the function: ${e.message}")
}

Real-time

import ai.fal.falclient.*

runBlocking {
    val app = "110602490-lcm-sd15-i2i"
    val authKey = "$FAL_KEY"

    val onMessage: (String) -> Unit = { message ->
        println("Received message: $message")
    }

    val onError: (Throwable) -> Unit = { error ->
        println("Error occurred: ${error.message}")
    }

    val webSocketConnection = WebSocketConnection(app, onMessage, onError)

    webSocketConnection.connect(authKey)

    val inputMap = hashMapOf(
        "prompt" to "an island near sea, with seagulls",
        "image_url" to "https://storage.googleapis.com/falserverless/model_tests/lcm/source_image.png",
        "seed" to 6252023,
        "sync_mode" to 1,
        "strength" to 0.8,
        "num_inference_steps" to 4,
    )

    while(!webSocketConnection.isConnected()) {
        delay(100)
    }

    webSocketConnection.send(Gson().toJson(inputMap))

    delay(1000)
 
    webSocketConnection.close()

}

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make to the Swift version of the client are greatly appreciated.

License

Distributed under the MIT License. See LICENSE for more information.