---

Multi-Cloud GPU Workflow Management Platform

A vendor-neutral, open-source multi-cloud GPU workflow management platform that integrates advanced workflow automation with GPU resource optimization to address ServiceNow limitations and multi-cloud challenges.

Features

Key Features of the Multi-Cloud GPU Workflow Management Platform:

• Vendor-Neutral & Open-Source: Flexible, adaptable to any cloud provider.
• Microservices Architecture: Modular, scalable infrastructure for efficient deployment.
• Advanced Workflow Automation: Low-code/no-code builder with AI-driven optimization.
• Seamless Multi-Cloud Integration: Unified GPU resource management across AWS, Azure, and GCP.
• GPU Virtualization Framework: Lightweight, cross-cloud resource sharing with security-first design.
• Dynamic Resource Allocation: ML-powered scheduling for cost-performance optimization.
• Comprehensive Analytics Engine: Real-time dashboards, predictive modeling, and performance insights.
• Enhanced Customization: API-first design enabling deep integrations with existing systems.
• AI-Powered Data Transformation: Real-time middleware for seamless legacy system integration.
• Cost-Effective Strategy: Intelligent caching and workload migration to optimize expenses.

How Does It Work?

Run Serverless AI Workloads

Add an endpoint decorator to your code, and you'll get a load-balanced HTTP endpoint (with auth!) to invoke your code.

You can also run long-running functions with @function, deploy task queues using @task_queue, and schedule jobs with @schedule:

from 7rainbow import endpoint


# This will run on a remote A100-40 in your cluster
@endpoint(cpu=1, memory=128, gpu="A100-40")
def square(i: int):
    return i**2

Deploy with a single command:

$ 7rainbow deploy app.py:square --name inference
=> Building image
=> Using cached image
=> Deployed 🎉

curl -X POST 'https://inference.beam.cloud/v1' \
-H 'Authorization: Bearer [YOUR_AUTH_TOKEN]' \
-H 'Content-Type: application/json' \
-d '{}'

Run on Bare-Metal Servers Around the World

Connect any GPU to your cluster with one CLI command and a cURL.

$ 7rainbow machine create --pool lambda-a100-40

=> Created machine with ID: '9541cbd2'. Use the following command to set up the node:

#!/bin/bash
sudo curl -L -o agent https://release.beam.cloud/agent/agent && \
sudo chmod +x agent && \
sudo ./agent --token "AUTH_TOKEN" \
  --machine-id "9541cbd2" \
  --tailscale-url "" \
  --tailscale-auth "AUTH_TOKEN" \
  --pool-name "lambda-a100-40" \
  --provider-name "lambda"

You can run this install script on your VM to connect it to your cluster.

Manage Your CPU or GPU Fleet

Manage your distributed cross-region cluster using a centralized control plane.

$ 7rainbow   machine list

| ID       | CPU     | Memory     | GPU     | Status     | Pool        |
|----------|---------|------------|---------|------------|-------------|
| edc9c2d2 | 30,000m | 222.16 GiB | A10G    | registered | lambda-a10g |
| d87ad026 | 30,000m | 216.25 GiB | A100-40 | registered | gcp-a100-40 |

Setting Up the Server

k3d is used for local development. You'll need Docker to get started.

To use our fully automated setup, run the setup make target.

make setup

Local DNS

This is required to use an external file service for mulitpart uploads and range downloads. Its optional for using the subdomain middlware (host-based URLs).

brew install dnsmasq
echo 'address=/cluster.local/127.0.0.1' >> /opt/homebrew/etc/dnsmasq.conf
sudo bash -c 'mkdir -p /etc/resolver'
sudo bash -c 'echo "nameserver 127.0.0.1" > /etc/resolver/cluster.local'
sudo brew services start dnsmasq

To use subdomain or host-based URLs, add this to the config and rebuild the Beta9 gateway.

gateway:
  invokeURLType: host

You should now be able to access your local k3s instance via a domain.

curl http://beta9-gateway.beta9.svc.cluster.local:1994/api/v1/health

Setting Up the SDK

The SDK is written in Python. You'll need Python 3.8 or higher. Use the setup-sdk make target to get started.

make setup-sdk

Using the SDK

After you've setup the server and SDK, check out the SDK readme .

Contributing

We welcome contributions big or small. These are the most helpful things for us:

• Open a PR with a new feature or improvement