Modal
Serverless GPU compute: a Python decorator puts a function on an accelerator, scales it from zero to thousands of containers and stops billing when it stops running — aimed at inference, fine-tuning and batch jobs rather than reserved clusters.
What it does
- GPU cloud
- Model inference
- Model hosting
Features 9
- FunctionsServerless compute that runs arbitrary code with a local-development feel and fans out over thousands of GPUs.
- API gatewaysTurns any function into an HTTPS endpoint that scales with traffic and idles at zero.
- Distributed filesystemA distributed filesystem for loading container images, model weights and datasets at full bandwidth.
- Globally distributed capacityRoutes workloads across 20+ clouds to the right hardware and region automatically.
- Automated fleet healthContinuous passive and active health checks on every node, with bad hardware drained and replaced.
- Enterprise security controlsWorkload isolation, configurable network and endpoint access, RBAC, audit logs and data residency controls.
- OpenTelemetry observabilityLogs and metrics for every function and sandbox, exportable to any OpenTelemetry provider.
- Cloud bucket mountsCloudBucketMount is a mutable volume that allows both reading and writing files from a cloud bucket.
- OIDC integrationUses automatically generated identity tokens to authenticate to external services.
How it compares
-
Replicate
Direct competitors include Baseten, which raised $150M Series D and focuses on mission-critical inference with dedicated deployments, and Replicate
Research report · 8 Oct 2026 -
Baseten
Direct competitors include Baseten, which raised $150M Series D and focuses on mission-critical inference with dedicated deployments
Research report · 19 Sep 2026 -
Together Inference
Specialized AI platforms like Modal, Together.ai, and Fireworks.ai
Official documentation · 19 Sep 2026 -
Fireworks AI
Specialized AI platforms like Modal, Together.ai, and Fireworks.ai
Official documentation · 19 Sep 2026 -
Serverless
Runpod vs Modal: Python-native serverless versus portable GPU compute
Official documentation · 19 Sep 2026 -
Pods
Runpod is the strongest all-round Modal alternative for teams that want on-demand GPU compute with more control than pure serverless allows.
Company disclosure · 13 Oct 2026
Sources
- Pricing
-
Plan tiers, included credits, per-second GPU/CPU/memory rates and volume pricing all read from the page 2026-09-04.
- Description
-
Official documentation · 8 Oct 2026
"In a few lines of code, we let you scale from zero to thousands of CPUs or GPUs." Matches the stored description; only the citation was missing.
- Constraints
-
Official documentation · 2 Oct 2026
Derived from the security facts on this page, from the same source.
- Security
-
Official documentation · 2 Oct 2026
Platform-wide: security, data-residency, region-selection, audit-logs, SCIM and RBAC docs. HIPAA BAA is Enterprise plan only and excludes Volumes v1, Images (most), Memory Snapshots and user code. SSO, SCIM and audit logs are Enterprise-plan features; SCIM is beta. ZDR unknown at platform level: only Inference endpoints are described as zero data retention. Training is from the Modal terms: without prior written consent Modal will not 'train any AI model using Customer Data' (this page adds that Modal will not access inputs, outputs or stored data). Customer Supplied Encryption Keys exist but are Alpha, so customer-managed-keys not recorded. Trust portal trust.modal.com is script-rendered.
- Sold within
-
Modal's pricing page offers "Get started" and "Sign up" and lists "Starter: $0 + compute / month", "Team: $250 + compute / month" and "Enterprise: Custom compute", with "$30 / month" of included compute on Starter.
- Interfaces
-
Official documentation · 8 Oct 2026
"Run pip install modal to install the modal Python package" -- a Python SDK and a CLI workflow for every product on the same account/meter; modal-notebooks additionally has a browser notebook UI.
- Intended for
-
"AI infrastructure that developers love"; also lists SOC2/HIPAA compliance and team controls for "teams of all sizes."
Also from Modal
-
Modal Inference Infrastructure service
Serve, scale and optimise model inference on Modal's runtime, with sub-second cold starts and autoscaling across regions.
-
Modal Training Infrastructure service
Managed training runs on Modal's fleet, configured in Python alongside the rest of a team's code; also gives instant access to GPU compute, including RDMA-enabled GPUs.
-
Modal Sandboxes Infrastructure service
Isolated, instantly-started containers for running untrusted or agent-generated code at scale — the primitive behind AI app-generation products.
-
Modal Notebooks Developer tool
Hosted notebooks backed by Modal's GPUs, for profiling and experimenting without provisioning a machine.
-
Modal Batch Infrastructure service
Batch execution of large jobs across Modal's fleet, described as one line of code on the product page.
Something wrong here? Send a correction — quote this product id: modal-platform.