Home

An open-source runtime that implements Modal's Python interface. Write the same code, run it on your own infrastructure.

import openmodal

app = openmodal.App("my-experiment")

@app.function(gpu="H100")
def train(config):
    ...

results = train.map(configs)

What works

• f.local(), f.remote(), f.map()
• GPU serving with auto scale-to-zero (KEDA)
• Custom images, secrets, retries, volumes
• Sandboxes for SWE agents (parallel creation, exec, file transfer)
• Local Docker provider — no cloud account needed
• GCP provider with spot GPUs (H100, A100, L4)
• AWS provider with EKS, Karpenter, KEDA
• Azure provider with AKS, ACR, KEDA
• CLI: openmodal run, deploy, stop, ps, logs, monitor, secret, setup
• openmodal monitor — live GPU/CPU/memory dashboard
• openmodal secret — manage named secrets locally
• Benchmark suite for sandbox performance testing
• CooperBench integration — one-line import swap (import openmodal as modal)
• Harbor integration for SWE-bench evaluations

Quick start

Local (Docker)GCPAWSAzure

pip install openmodal
openmodal --local run examples/hello_world.py

Just needs Docker installed. No cloud account, no setup.

pip install openmodal
openmodal setup gcp
openmodal run examples/hello_world.py

The setup wizard handles authentication, API enablement, and tool installation.

pip install "openmodal[aws]"
openmodal setup aws
openmodal --aws run examples/hello_world.py

Creates an EKS cluster on first run (~15 min one-time).

pip install openmodal
openmodal setup azure
openmodal --azure run examples/hello_world.py

Creates an AKS cluster on first run (~5 min one-time).

How it compares to Modal

The only difference is the import line:

- import modal
+ import openmodal

For hundreds of additional examples, see the Modal examples gallery — the same code works with OpenModal.

License

Apache-2.0