AI Tool
Mastra pricing, features, company info, and alternatives
A factual product page for Mastra's TypeScript agent framework and platform.
Last updated April 2026 · Pricing and features verified against official documentation
Pricing
Current public pricing tiers on file for Mastra, last verified Apr 23, 2026.
Platform Starter
$0 / month
Unlimited users and deployments, plus Studio, 100,000 observability events, 24 hours of CPU uptime, and 10GB of data egress.
Platform Teams
$250 / team/month
Multiple teams, custom SSO, SOC 2 documentation, 250 hours of CPU time, and 100GB of data egress.
Platform Enterprise
Custom
RBAC, support SLA, dedicated support engineer, custom CPU time, and custom data egress.
Memory Gateway Starter
$0 / month
100K memory tokens, $10/1M add-on tokens, 250MB retrieval storage, 15-day stale thread retention, and an initial $5 credit.
Memory Gateway Teams
$250 / team/month
1M memory tokens, $10/1M add-on tokens, 1GB retrieval storage, 6-month stale thread retention, and bring-your-own-key support.
Memory Gateway Enterprise
Custom
RBAC, support SLA, dedicated support engineer, and custom storage or retention.
What You Can Do With It
The main capabilities that shape how people use Mastra today.
Open-source TypeScript framework for agents, workflows, RAG, observability, tools, and MCP support.
Mastra Studio adds traces, logs, metrics, scorers, and evals for production observability.
Mastra Server exposes agents and workflows as APIs or bundles them with an application.
Memory Gateway provides managed agent memory for Mastra and other frameworks.
Best For
Who Mastra is most clearly built for.
TypeScript teams that want an open-source agent framework with a first-party platform around it.
Builders shipping agentic apps that need deployment, observability, and auth controls in one stack.
Teams that want managed memory and evaluation tooling alongside their runtime.
Company
Leadership and company context for Mastra.
CEO
Sam Bhagwat
Founders
Sam Bhagwat, Shane Thomas, Abhi Aiyer
Investors
Spark Capital, Gradient, basecase, Y Combinator
Headquarters
San Francisco, CA, USA
Platforms
Where you can use Mastra today.
Web
CLI
TypeScript package
Self-hosted
Integrations
Notable connected tools and ecosystem hooks for Mastra.
Next.js
Express
Hono
WorkOS
Langfuse
Braintrust
Privacy Notes
Publicly stated data-handling notes that matter when evaluating Mastra.
Mastra's privacy policy says it collects personal information and log data to provide and improve the service.
The policy says third-party service providers may access personal information only to perform assigned tasks.
The policy says Mastra uses commercially acceptable means to protect personal information, but no internet transmission or storage method is 100% secure.
Access
How to integrate or build around Mastra.
Public API
Yes
Docs
Available
Alternatives
Other tools worth considering alongside Mastra.
Multi-agent AI platform for building crews, flows, and enterprise automations.
AI observability and evaluation platform for tracing, scoring, and improving production AI applications.
Serverless AI infrastructure platform for inference, training, batch jobs, sandboxes, and notebooks.
AI gateway and observability platform for production LLM apps.
Product Snapshot
Mastra is an open-source TypeScript framework for AI-powered applications and agents. The current product family includes the framework, Studio, Server, and Memory Gateway.
What You Can Do With It
- Build agents, workflows, RAG pipelines, and MCP-connected tools in a TypeScript-first framework.
- Use Studio for traces, logs, metrics, scorers, evals, datasets, and experiments around production behavior.
- Deploy agents and workflows through Server as APIs or bundle them directly with your application.
- Add managed memory through Memory Gateway for Mastra or other framework-based agent stacks.
Why It Stands Out
It combines an open-source runtime with first-party observability, deployment, auth, and memory products rather than treating those as third-party add-ons.
Tradeoffs To Know
- Mastra’s pricing is split across the platform and Memory Gateway products, so total cost depends on which surfaces you adopt.
- The product family spans framework, Studio, Server, and memory services, which is broader than a single-purpose observability or deployment tool.
- Some governance controls, such as custom SSO and RBAC, are tied to higher tiers or specific product surfaces.