Plans & Pricing
Guide to pricing and model usage for Free, Pro, and Enterprise plans
Upgrade your plan and view your usage on your account page
Available Plans
- 10 active agents
- 2 agent templates
- 50 premium requests
- 500 standard requests
- 1 GB of storage
- 1,000 active agents
- 20 agent templates
- 500 premium requests
- 5,000 standard requests
- 10 GB of storage
- 100,000 active agents
- 100 agent templates
- 5,000 premium requests
- 50,000 standard requests
- 100 GB of storage
Enterprise (contact us)
- Up to agents & storage
- Custom model deployments
- SAML/OIDC SSO authentication
- Role-based access control
- BYOC deployment options
Understanding Agents vs Templates
In Letta Cloud, you can use agent templates to define a common starting point for new agents. For example, you might create a customer service agent template that has access a common set of tools, but has a custom memory block with specific account information for each individual user. Read our templates guide to learn more.
Understanding Requests
Your Letta agents use large language models (LLMs) to reason and take actions. These model requests are what we count toward your monthly requests quota.
Standard vs Premium Model Requests
Standard models (GPT-4o mini, Gemini Flash, etc.) are faster and more economical. They’re ideal for simple tool calling and basic chat interactions.
Premium models (GPT-4.1, Claude Sonnet, etc.) offer enhanced capabilities for complex agentic tasks. They excel at multi-step tool sequences and tasks requiring advanced reasoning.
Some high-powered models (like o1 and o3) are available exclusively through usage-based pricing.
How Requests Are Counted
Each agent “step” or “action” counts as one model request. Complex tasks (such as deep research) may require multiple requests to complete. You can control request usage via tool rules that force the agent to stop on certain conditions.
Quota Refresh
Request quotas refresh every month. Free plan quotas refresh on the 1st of each month. Pro plan quotas refresh at the start of your billing cycle. Unused requests do not roll over to the next month.
Usage-based Pricing
If you are on the Pro plan, you can enable usage-based pricing to allow you to continue to make model requests after you’ve exceeded your request quota. Unused credits purchased roll over on each billing cycle.
Usage-based billing can be enabled by adding credits to your account under your account settings page. See a full model list and pricing here.
Enterprise Plans
For organizations with higher volume needs, our Enterprise plan offers increased quotas, dedicated support, role-based access control (RBAC), SSO (SAML, OIDC), and private model deployment options. Contact our team to learn more.