Open-Source Agentic AI Infrastructure

Where intelligence
grows.

Modular, composable tools for agentic AI systems. Orchestrate, evaluate, and refine intelligence across models, providers, and workflows — with clarity, transparency, and control.

The Petal Ecosystem

Five petals. One living architecture.

Go Graph RuntimeView on GitHub
LLM → Router → Tool → Memory

Compose, execute, and serve agentic workflows as living graphs. Wire LLMs, tools, dynamic routing, schedules, and webhooks together without locking into rigid orchestration frameworks.

Composable DAG executionIris provider adapterSchedules & webhooks

Philosophy

AI should feel organic.

Petal Labs builds modular, composable tools for agentic AI systems. Our mission is to empower developers to orchestrate, evaluate, and refine intelligence across models, providers, and workflows — with clarity, transparency, and control.

We believe AI should feel organic: systems that grow, adapt, and interconnect rather than lock you into rigid pipelines. Each tool in the Petal ecosystem is designed like a petal in a larger bloom — independent, purposeful, and capable of forming something extraordinary when combined.

01

Independent & Composable

Adopt a single petal like Iris or Cortex on its own, or weave all five together into a complete agentic architecture without framework lock-in.

02

Built for Clarity & Control

Every model call, graph branch, memory lookup, and trace span is inspectable and deterministic — replacing black-box magic with real engineering visibility.

03

Organic by Design

AI systems evolve fast. Our tools are lightweight, provider-agnostic primitives designed to adapt as models, protocols, and workflows grow.

Grow with Petal Labs.

Interested in collaborating, early access to Tahiti, or building with the Petal ecosystem? Leave your email and we'll be in touch.