Skip to main content
This page collects everything you need beyond the core documentation: the open-source libraries you install, the API references you look up, and the community channels where you can get help. Whether you are calling models from Python or JavaScript, deploying your own with the CLI, or integrating fal into a larger system with the Platform APIs, the links below will get you there. All fal libraries are open source and available on GitHub. The platform itself offers a model gallery, a dashboard for managing your account, and a sandbox for comparing models side by side. If you get stuck, the Discord community and blog are good places to start.

​
Open-Source Packages

PackageInstallWhat it doesSource
falpip install falDeploy models with the serverless runtime, CLI (fal deploy, fal run), and toolkit (Image, File, download_file)fal-ai/fal (Apache-2.0)
fal-clientpip install fal-clientCall models from Python with subscribe, submit, upload_filefal-ai/fal (Apache-2.0)
@fal-ai/clientnpm install @fal-ai/clientCall models from JavaScript with the same API as the Python client, with TypeScript typesfal-ai/fal-js (MIT)
@fal-ai/server-proxynpm install @fal-ai/server-proxyServer-side proxy for web apps that keeps your API key off the client (Next.js, Express)fal-ai/fal-js (MIT)
flashpackpip install flashpackFast model loading with high-throughput tensor loading for PyTorch (up to 25 Gbps)fal-ai/flashpack (MIT)

​
Platform

Model Gallery

Browse 1,000+ production-ready AI models (docs)

Dashboard

Manage apps, keys, billing, and monitoring (analytics, events, errors)

Sandbox

Compare models side by side in your browser (docs)

Status Page

Real-time platform health and incident history

​
Documentation

The docs are organized into three main areas depending on how you use fal. Model APIs covers calling existing models, Serverless covers deploying your own, and Compute covers dedicated GPU instances for training or custom workloads.

Model APIs

Call 1,000+ models with simple API calls: inference, streaming, real-time

Serverless

Deploy your own models: environment setup, endpoints, scaling, observability

Compute

Dedicated GPU instances with SSH access for training and custom workloads

Examples

Step-by-step tutorials for image, video, and audio generation

Changelog

Platform updates and new features

OpenAPI Spec

Machine-readable API specification for building custom integrations

​
API References

For programmatic access beyond the client libraries, fal exposes a REST API for managing apps, files, and keys, a full CLI reference, and the Python SDK docs.

Platform APIs

REST API for managing apps, files, metrics, and keys

CLI Reference

Full command reference for the fal CLI

Python SDK

Python SDK API reference

​
Client Libraries

fal has official client libraries for Python and JavaScript, plus community-maintained clients for Swift, Kotlin, and Dart. Each library handles authentication, request queuing, and file uploads so you can focus on your application.

Python

pip install fal-client

JavaScript

npm install @fal-ai/client

Swift

iOS client

Kotlin

Android / JVM client

Dart

Flutter client

​
Community and Support

The fal community is active on Discord, where you can ask questions, share projects, and get help from the team and other developers. The blog covers product updates and engineering deep dives.

Discord

Join the community, ask questions, share what you’re building

Blog

Product updates, tutorials, and engineering deep dives

Contact Sales

Enterprise plans, dedicated support, and custom solutions