Hey there,
We’re updating how pricing works for Replit Agent to better reflect the effort the Agent puts forth in addressing user requests.
Previously, the Agent charged $0.25 per checkpoint. This meant small edits would cost $0.25, while larger tasks could trigger multiple $0.25 checkpoints.
With our new Effort-Based pricing model, small changes still result in a single checkpoint, but that checkpoint will typically cost less than $0.25. Larger or more complex tasks are no longer split across multiple $0.25 checkpoints. Instead, they will be bundled into one checkpoint, which may cost more than $0.25 but reflects the total effort involved.
This means you’ll pay less for simple requests and may pay more for complex requests—ensuring pricing aligns better with the actual work the Agent performs on your behalf.
The new pricing model is effective immediately for new Replit users and will start rolling out to existing Core and Teams subscribers starting July 1st.
New pricing model Effort-Based pricing: Cost will reflect the true scope of the Agent’s work More meaningful checkpoints per request: Checkpoints capture the final output of the request only No intermediate checkpoints: This reduces noise and makes it easier to track progress Simple tasks may cost less than $0.25, more complex tasks may cost more than $0.25
How the pricing will appear in the Replit Agent user interface
Why we're changing checkpoint pricing
The original $0.25-per-checkpoint pricing was designed for the early version of Replit Agent, which mostly handled short, simple tasks that took a consistent amount of time. As the Agent has evolved, it’s now capable of solving both more complex, longer-running tasks (like self-debugging errors) and smaller tasks more quickly. As a result, the time and effort involved in addressing user requests can vary significantly from request to request.
Reflecting the Agent’s evolving capabilities, we are introducing two new controls to adjust the level of effort the Agent uses—available on a per-request basis. These new controls will start rolling out to new users starting later today and to existing users starting July 1st. High power model: Increases the Agent’s intelligence by using a more capable AI model Extended thinking: Instructs the Agent to take more time for deeper reasoning and more thorough solutions
For more complex tasks or tough challenges, using a more powerful and thorough version of the Agent can be a smarter use of your time and budget. It’s like calling in an expert when the problem demands it.
With Effort-Based pricing, you stay in control—choosing when to spend extra resources for a more powerful Agent—and you’ll only pay for the actual effort required to complete each request.
Marching towards a more agentic future
The Agent is getting smarter, with enhanced capabilities to tackle more complex, nuanced, and long-running tasks. Effort-Based pricing lays the groundwork for a more agentic future—one where Agent runs longer autonomously, building and testing more of your app with minimal user input.
Keep building, The Replit Team