Skip to main content

Runs

Scale to zero

Ephemeral batch compute

Run batch workloads as ephemeral containers that start instantly and terminate on exit. Spawn dozens of parallel runs from your server code, mount shared volumes for feature caches, and get exit codes and logs back — no infrastructure to manage. Pay only for the compute you use.

Overview
Pricing
Usage
Docs
Examples

Key Features

Parallel Execution

Spawn N runs simultaneously for fan-out workloads

Run to Completion

Job runs until exit — no idle timeouts, no heartbeats

Shared Volume Mounts

Mount ReadWriteMany volumes for shared feature caches

Resource Limits

Set CPU and memory per run for predictable cost

Logs Capture

stdout/stderr captured and stored on completion

Image Security Scanning

Images scanned for vulnerabilities before execution

Use Cases

ML model training and evaluation
ETL pipelines and data processing
Batch report generation
File transformation and transcoding
Parallel hyperparameter search

At a Glance

Pricing
$0.00004
per vCPU-second
Free Tier
100K vCPU-seconds/month
Features
6 included

Ready to get started?

Check out the docs tab for quick start guides, code examples, and API reference.

View Documentation