“Modal was the only infrastructure provider that enabled us to reliably run tens of thousands of app creation sessions in an instant. We’re excited to build with them for the long term.”
“Sandboxes are one of the most important building blocks for RL. Modal was clearly very flexible, structured in a way where we could build these complex environments, and really focused on performance and reliability.”
“It would have been really hard to build Inspect without Modal. A lot of the complexity is hidden behind just being able to spin up sandboxes very quickly with a lot of services and data, and effectively have infinite of them at any given time.”
Modal's infrastructure scales to the parallelism that production agent systems and RL training actually demand.
Bring your dependencies, your base image, your setup scripts. Sub-second scheduling and strong cold-start performance on custom images
Built on a multi-cloud, multi-region foundation and Modal’s custom AI-native infrastructure means deep capacity pools and no waiting.

Granular snapshotting on any image
Bring your own image or build one in Python, scale resources as needed, and leverage state-of-the-art GPUs like H100s & A100s for high-performance computing.
Integrated, distributed storage
Mount modal.Volume across sandbox runs to share artifacts, persist outputs, and give long-horizon agent workflows a durable filesystem that survives sandbox restarts.
Manage context across sandboxes
Give sandboxes a shared memory layer. Pass context between runs, coordinate across parallel workloads, and store lightweight state without standing up a database.
Networking
Expose ports from running sandboxes with tunnels, inject credentials with Secrets, and trigger sandbox creation from anywhere in your app with Functions, all on the same platform.

Only pay for what you use. Burst up to what you need without over-allocating CPU or memory in advance.
CPU
Physical core
(2 vCPU equivalent)
$0.00003942 / core / sec
Memory
$0.00000672 / GiB / sec
GPU
See standard pricing
