> For the complete documentation index, see [llms.txt](https://docs.oncompute.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.oncompute.ai/ocean-orchestrator/editor.md).

# Ocean Orchestrator

Run affordable GPU AI jobs from your editor with a one-click workflow and pay-per-use mechanism.

**Ocean Orchestrator** lets you submit a containerized compute job to remote nodes, monitor it, and automatically pull the results back. It uses **Ocean Compute-to-Data (C2D)**, meaning the job runs in an isolated container near the data and only the outputs are returned. This gives you a low-friction way to run batch jobs without spinning up or managing servers.

Ocean Network coordinates those remote runs across a distributed set of GPU nodes, handling orchestration behind the scenes so pay-per-use compute jobs stay usable even as more nodes and workloads join.

Works in **VS Code**, **Cursor**, **Antigravity**, and **Windsurf**.

### Features

* **One-Click Job Runs**: Run a job without spinning up servers. Create a project, press Start Compute Job, and receive the outputs in your folder. Supports Python and JavaScript projects with built-in templates and dependencies.
* **Pay Per Use Compute**: Start with free compute for quick tests, then switch to paid compute jobs when you need more resources.
* &#x20;**Compute-to-Data**: Your code runs in an isolated container, and only results are returned, so data stays sealed.
* **Remote AI Compute**: Run embeddings, inference, data cleanup, batch processing, and other containerized workloads without provisioning servers.
* **Real Time Monitoring**: Track job status and view logs directly in your editor via the Output console.
* **Automatic Results Retrieval**: Outputs and logs are saved to your results folder as soon as the job completes.

### Getting Started

1. **Install**: Install **Ocean Orchestrator** from your favorite extension marketplace (VS Code, Cursor, Antigravity, Windsurf). If your editor does not support the VS Code Marketplace, use the Open VSX listing.&#x20;

<figure><img src="/files/vK6RHFitSShQqaiBvDlq" alt=""><figcaption></figcaption></figure>

1. **Open Panel**: Click the Ocean icon in the Activity Bar to open the Ocean Orchestrator panel.
2. **Create Project**:
   * Click "Create a new project folder".
   * Choose a parent directory and name your project (default: `new-compute-job`).
   * Select your language: **Python**, **JavaScript**, or **Custom Container**.
3. **Explore Structure**:
   * Algorithm file (`.py` or `.js`)
   * `Dockerfile` with environment setup
   * Dependencies file (`requirements.txt` or `package.json`)
   * `.env` secrets file
4. **Run**: Click **Start FREE Compute Job**.
5. **Monitor**: Watch job status and real-time logs in the Output console.
6. **Results**: Check outputs and logs in your project's `results` folder.

### Extension Layout

Ocean Orchestrator adds a dedicated Ocean section to the activity bar. From there, you can:

* Optionally select a dataset file.
* Create a new compute project or select an existing one.
* View available compute resources under **Setup**.
* Configure compute settings under **Configure Compute**.
* Start free or paid compute jobs.

### Starting a Compute Job

1. Create a new project folder or select an existing one.
2. Review your algorithm, Dockerfile, and dependencies.
3. Click **Start Free Compute Job** or switch to paid for more resources.
4. Monitor job status and real-time logs in the Output console.
5. Check outputs in the `results` and `logs` folders.

### Advanced Setup

<figure><img src="/files/VBjrXmNdfB0JJCpULW0b" alt=""><figcaption></figcaption></figure>

#### Custom Docker Image

Use your own docker image if you are not using a Dockerfile in the project folder.

#### Compute Resources

* **Free Compute**: Uses minimal resources for testing.
* **Paid Compute**: Runs on demand and charges per run based on resources, time, and environment selection. See available tiers under **Setup**.

#### Node Status Check

Use **Check** under **Setup** to verify node availability before running a job.

### Troubleshooting

* **Job cannot start**: Check the node status under **Setup**, then press **Check**.
* **Not enough funds**: Switch to free compute or top up your account.
* **General issues**: Check logs in the Output console. Logs are also saved in your project folder under `logs`.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.oncompute.ai/ocean-orchestrator/editor.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
