runcrate

n8n community node for Runcrate: chat with 200+ open models, list GPU offers, launch GPU instances and run GPU jobs.

Package Information

Released: this month
Downloads: 0 weekly / 0 monthly
Latest Version: 0.1.0
Author: Aeonmind LLC

Available Nodes

Documentation generating…

Runcrate icon
Runcrate
Run open models and launch GPUs on Runcrate

Documentation

n8n-nodes-runcrate

This is an n8n community node for Runcrate, a GPU cloud for AI developers and the agents they run. Use it in workflows, or give it to an n8n AI Agent as a tool, to call 200+ open models through Runcrate's OpenAI-compatible API, find GPU offers, run commands on GPUs and launch GPU instances.

Installation · Operations · Credentials · Compatibility · Usage · Resources

Installation

Follow the installation guide in the n8n community nodes documentation. In n8n: Settings > Community Nodes > Install, then enter n8n-nodes-runcrate.

Operations

Resource Operations
Chat Complete: send a prompt (and an optional system message) to any chat model, with a model dropdown loaded live from the API
Model Get Many: every inference model with its prices and modalities
GPU Type Get Many: launchable GPU configurations with region and hourly price, filterable by GPU type, count and region
Job Create (run a shell command on a fresh GPU machine that stops billing when the command ends), Get, Get Many, Get Logs, Cancel
Instance Create (launch a GPU machine with SSH), Get, Get Many, Get Status, Delete
Account Get Balance

Jobs always carry a max runtime, and instances can be given one too, so an automation never leaves a GPU running.

Credentials

  1. Create an API key in the Runcrate dashboard. Keys start with rc_live_.
  2. In n8n, add a Runcrate API credential and paste the key. The credential test reads your balance.

Runcrate is pay as you go: inference is billed per token, image or second, and GPUs by the hour (most bill by the second) from a prepaid balance.

Compatibility

Built with @n8n/node-cli 0.53 and tested on n8n 2.44.

Usage

  • Chat model in a workflow: Runcrate > Chat > Complete, pick a model such as deepseek-ai/DeepSeek-V3.2, map the prompt from an earlier node. Turn off Simplify Output for the full OpenAI-style response.
  • GPU job from a schedule or webhook: Runcrate > Job > Create with a command like python train.py, a GPU type such as A4000, RTX4090 or H100, and a max runtime. Poll Job > Get until the status is succeeded, then read Job > Get Logs.
  • AI Agent tool: the node is usable as a tool, so an n8n AI Agent can check GPU prices or start a job itself.
  • OpenAI-compatible: n8n's own OpenAI Chat Model node also works with Runcrate: set the base URL to https://api.runcrate.ai/v1.

Resources

Version history

  • 0.1.0: Chat, Model, GPU Type, Job, Instance and Account resources.

Discussion