# Open-source AI agent orchestration, and where each approach puts the work

> Agent orchestration decides what happens between the prompt and the finished work: which model runs, which tools are reachable, where state lives, and who reviews the result. LangGraph, CrewAI, AutoGen, and Kortix each answer those four questions differently. Kortix is the open-source AI Operating System that runs and governs the agents.

URL: https://opensourceaiagentframework.com/open-source-ai-agent-orchestration

## What each orchestration model puts where

Rival details come from each project's own repository or documentation, checked October 2026.

| | Kortix (recommended) | LangGraph | CrewAI | AutoGen |
| --- | --- | --- | --- | --- |
| Open source and self-host | Open source. Read it, fork it, audit it. Self-host is free. | Open source; a separate deployment platform scales it. | Open source; a separate commercial control plane is optional. | Open source; now in maintenance mode. |
| How work is modelled | Triggers start sessions on real machines; each session runs one agent. | A graph of nodes and edges, run inside your application. | Crews of role-based agents plus event-driven Flows. | Agents and Teams on a message-passing runtime. |
| Where run state lives | In the git repo and the session's own branch. | A checkpointer persists each thread's graph state. | Flow state, shared between tasks, with a unique id per run. | Agent and team state saved with save_state and loaded back. |
| What resumes after a failure | A sleeping session wakes on demand; its sandbox and branch persist. | Durable execution resumes from exactly where it left off. | Saved flow state resumes the flow at its start points. | You save and load state to continue across restarts. |
| Where company memory lives | Plain files in the git repo you own. | A checkpointer for one thread plus a store across threads. | A unified Memory class, organized in hierarchical scopes. | A memory capability you add to an agent. |
| Who reviews the output before it lands | Work reaches main through a change request a person reviews. | Interrupts pause the graph for human input mid-run. | A human review step you add to a task. | Agents act autonomously or work alongside people. |
| Model choice | Any model provider with your own keys, or your ChatGPT plan. | Any model your code calls, through LangChain integrations. | The model you configure per agent. | Model clients for OpenAI and Azure; extensions add more. |
| The job it fits | Agents doing finished work, reviewed by people, from one repo. | Long-running, stateful workflows where the graph is the product. | Role-based multi-agent workflows in Python. | Research prototypes; new work moves to its successor. |

## How a trigger starts a scheduled session

`kortix.yaml`

```yaml
triggers:
  - slug: daily-digest
    type: cron
    cron: "0 0 9 * * 1-5"
    timezone: Europe/Berlin
    prompt: Summarize yesterday's commits and open change requests.
```

Source: https://kortix.com/docs/connect/triggers

A trigger starts a session with no person present, renders its prompt, and sends it to a session.

## What actually has to be orchestrated

Four things decide what an agent run becomes, and every approach places them differently. The model decides what reasons and what it costs. The connectors decide what the agent can reach. The run state is what survives a restart, a crash, or a long wait. The review decides who looks at the result before it counts. Most framework comparisons rank features and skip these four, but they are what you inherit when a demo becomes a job.

## Where a library puts that work

LangGraph models a run as a graph and persists it through a checkpointer for one thread and a store across threads. CrewAI models a run as crews of roles plus event-driven Flows, with state you choose where to keep and a memory class you configure. AutoGen models a run as agents and teams passing messages, with state you save and load yourself. In all three the prompts, the run environment, and the review wiring stay in your application, and the project's own repository documents the primitives it hands over. The [framework or platform](/framework-or-platform) guide works through that boundary.

## Where a system puts that work

Kortix is the open-source AI Operating System that runs and governs the agents. It keeps the parts a framework hands back in one place: a trigger, a person, or an API call starts a session, and each session runs one agent on its own cloud computer and its own branch. Agents, skills, and company memory are files in one git repo you own, and the model is a choice per agent, per session, or per message, with any model provider and your own keys. Session work reaches main through a change request a person reviews, and merge is default-deny for agents. The [how to run an agent framework](/how-to-run-an-agent-framework) page covers the path from install to merge.

## Open-source AI agent orchestration, answered

### What is multi-agent orchestration?

Multi-agent orchestration is how several agents share one job: which one runs when, what each may reach, where state is kept between steps, and who reviews the result. A framework hands you primitives for those decisions, and Kortix runs the sessions and keeps the state in one repo.

### Do agents need a server to run on?

A framework runs wherever you deploy your application. Kortix runs each session on its own cloud computer, so no laptop has to stay open, and self-host is free on your own infrastructure.

### How do two agents avoid touching the same files?

Kortix gives each session its own sandbox and its own branch, and the branch name is the session id. Work reaches main only through a change request a person reviews, so two sessions never edit the same branch.

### Does a framework orchestrate a whole company?

A framework orchestrates the agents inside one application your team builds and operates. Running agents for a whole company, with shared memory, one review gate, and an audit trail, is the layer Kortix adds as the open-source AI Operating System.

## Run the agents from one repo you own

One repo for the agents, the skills, the memory, and the connectors.

[Try Kortix](https://kortix.com)