flyteorg/flyte

Project overview

Flyte addresses the complex challenge of orchestrating ML pipelines and agent tasks at scale through a pure-Python development model, backed by its standing as a graduated project of the LF AI & Data Foundation.

Project type

Workflow

Use cases

Automation

Deployment

Refer to project documentation

License

Apache-2.0

Best for

• Developers, AI engineers, and data teams who want to build and orchestrate resilient machine learning pipelines and agent tasks using pure Python.

Key capabilities

• Build and run machine learning pipelines and agent tasks at scale using pure Python, with support for local execution.
• Provides a rich local development experience via an installable terminal user interface.

Getting started

• Flyte is easy to set up and installs via pip using standard Python tooling. The first success path is running uv pip install flyte.

Evidence and sources

• GitHub project description: Dynamic, resilient AI orchestration. Coordinate data, models, and compute as you build AI workflows.
• README: Reliably orchestrate ML pipelines, models, and agents at scale — in pure Python.
• README: uv pip install flyte
• README: @env.task def calculate(x: int) -> int: return x * 2 + 5
• README: from fastapi import FastAPI import flyte from flyte.app.extras import FastAPIAppEnvironment

Reviewed at: 2026-07-18T06:38:08Z