24 Sep (Tomorrow)Building a Classifier in Mastra with Jev

Mastra Server - Deploy your Mastra agents and workflows

Get StartedDocumentation

Ship Your Mastra Agent in One Command

Push code, get a URL. Routing, scaling, and infrastructure are handled for you.

The Natural Home for Mastra Agents

Deploy on a platform purpose-built for Mastra. Zero config for agents, workflows, memory, and telemetry.

Deploy from the CLI

Stay in your terminal. One command takes your local agent to a live endpoint.

Scalable API

Expose agents and workflows as API endpoints automatically. Call them from any service, at any scale.

Connect to Any Front-End

Plug the standard API into React, mobile, Slack, or whatever your users already live in.

Start building with a free account, or explore our pricing for more

Full Pricing

Starter

Free for everyone

$0/ month
  • 100K observability events+ $10/100K
  • 24 CPU hours+ $0.35/hr
  • 15 days of data retention
  • Unlimited users, deployments, and projects
Get started for free

Teams

For growing teams

$250/ month
  • 1M observability events+ $8/100K
  • 250 CPU hours+ $0.25/hr
  • 6 months of data retention
  • Multiple teams, SSO, and SOC 2 docs
Get Teams

Enterprise

For teams at scale

Custom pricing

Custom volume and retention, with RBAC, audit logs, support and uptime SLAs, and a dedicated support engineer.

Contact sales

Mastra is powering the best AI teams

Case studies

Medusa Builds E-commerce Agents with Mastra

Read story

How WorkOS Went From Evaluating Mastra to Teaching It and Using in Prod in ~1 Month

Read story

How Replit's Agent 3 Spins Up Thousands of Mastra Agent Sandboxes Weekly

Read story

Frequently asked questions

What is Mastra Server?

Mastra Server takes your Mastra agents from local development to a live API endpoint in one command. Deploy on infrastructure that understands agents, workflows, memory and telemetry with zero configuration. Mastra Server exposes your agents and workflows as API endpoints automatically, handles routing and scaling and plugs into React, mobile, Slack or whatever your users already live in. Secure endpoints with JWT, Clerk, Supabase, Firebase, Auth0 or WorkOS. Your agents are part of your infrastructure.

How do I deploy to Mastra Server?

Mastra Server deploys with a single mastra server deploy command that creates a live endpoint with routing, scaling and infrastructure handled for you. The CLI outputs the URL of your deployed Server instance on a successful deploy.

How does Mastra Server expose agents as API endpoints?

Mastra Server automatically exposes all registered agents and workflows as REST API endpoints when you deploy. Explore all available endpoints in the OpenAPI specification at /api/openapi.json or test them interactively in the Swagger UI at /swagger-ui. Call agent and workflow endpoints from any service, at any scale.

Can I use Mastra Server with my existing HTTP framework?

Mastra Server supports adapters for Express, Hono and other HTTP servers. Build adapters for frameworks not officially supported. The Mastra Client SDK provides a type-safe client for calling agents, workflows and tools from browser or server environments.

What authentication providers does Mastra Server support?

Mastra Server supports JWT, Clerk, Supabase, Firebase, Auth0 and WorkOS for securing agent endpoints. Configure authentication by adding an auth provider to your server configuration. Use middleware to intercept requests for authentication, logging, CORS or injecting request-specific context.

How does Mastra Server handle stream data redaction?

Mastra Server redacts system prompts, tool definitions, API keys and similar data from every streaming response before sending it to clients. Stream data redaction is enabled by default for both the generated server and server adapters.

Start building today.

Get StartedDocumentation