🚀 Quick Start

🌊 Floom is an AI Gateway that connects AI to your apps seamlessly & scalably in less than 🕒 5 minutes.

Generate, edit, detect or extract 🔤 text, 👩🏽‍💻 code, 🎧 audio, ▶️ video, 🖼️ images, 📄 documents and more.

📢 Follow for updates!

⭐ Star us on GitHub | Follow us on X, LinkedIn & Discord!

Hosting

Floom is tailored to every use case, offering flexible deployment options:

• Floom Cloud ☁️: Seamlessly host your AI Pipelines on our SaaS platform for immediate access and management.
• Docker Container 🐳: Deploy on-prem or in a private cloud for full control and customization.

Let's go!

1. Define your first Text Generation AI Pipeline:

   kind: floom/pipeline/1.2
   
   pipeline:
    name: my-first-pipeline
   
    prompt:  
     template: 
      package: floom/prompt/template/default
      system: "You are a nice assitant"

   Save this pipeline configuration YAML to 'my-first-pipeline.yml'.

2. Deploy

   Install Floom CLI:

   Mac

   brew install floom

   Windows

   winget install floom

   Linux

   apt-get install floom

   Choose your preferred hosting:

   ☁️ Floom Cloud

   floom deploy my-first-pipeline.yaml

   🐳 Docker

   # Install Floom Docker
   curl -O https://get.floom.ai/docker-compose.yml
   docker-compose up -d
   
   # Deploy pipeline
   floom deploy -endpoint=<YOUR-FLOOM-ADDRESS> -f my-first-pipeline.yaml

3. Run your AI Pipeline from your app

   Install via package managers:

   Python

   pip install floom

   Node

   npm install floomai

   .NET

   nuget install floom

   Java

   Add dependency to Maven:

   <dependency>
      <groupId>ai.floom</groupId>
      <artifactId>java-sdk</artifactId>
      <version>[version]</version>
   </dependency>

   💡Replace [version] with the desired version number: i.e. 1.0.2

   Golang

   go get github.com/FloomAI/FloomSDK-Go

   Rust

   cargo install floom

   Or directly Download Floom SDK for your preferred framework.

   
   Run your pipeline:

   Python

   from floom import FloomClient
   
   floom_client = FloomClient()
   
   floom_client.run(
      pipeline="my-first-pipeline", 
      input="Your query here"
   )

   Node

   import { FloomClient } from '../FloomClient';
   
   const floomClient = new FloomClient();
   
   const response = await floomClient.run(
      "my-first-pipeline",
      "Your query here",
   );

   .NET

   var floomClient = new FloomClient();
   
   var response = await floomClient.Run(
         pipelineId: "docs-pipeline",
         input: "How do I reset the oil alert in my dashboard?"
   );

   Java

   import ai.floom.FloomClient;
   
   FloomClient floomClient = new FloomClient();
   
   FloomResponse response = floomClient.run(
         "my-first-pipeline",
         "Your query here"
   );

   Golang

   import (
      "fmt"
      "github.com/FloomAI/FloomSDK-Go"
   )
   
   floomClient := floom.NewFloomClient()
   
   response, err := floomClient.Run(
      "my-first-pipeline", 
      "Your query here"
   )

   Rust

   extern crate floom;
   
   use floom::FloomClient;
   
   let mut floom_client = FloomClient::new();
   
   let response = floom_client.run(
      "my-first-pipeline",
      "Your query here",
   );

That's it!

By following these steps, you’ll have Floom integrated and ready to propel your AI development to new heights.

Learn about creating new AI Pipelines.