GitHub Copilot has introduced important changes to its Individual subscription plans to ensure a reliable and consistent experience for current users. These adjustments include pausing new sign-ups, implementing tighter usage limits, and modifying available AI models. The updates address the significant increase in computational demands driven by agentic workflows—where long-running, parallel tasks consume more resources than earlier plans were designed to handle. Below we answer common questions about these changes, why they were made, and what they mean for you.

Why are you pausing new sign-ups for GitHub Copilot Pro, Pro+, and Student plans?

We have temporarily paused new sign-ups for GitHub Copilot Pro, Pro+, and Student plans to better serve our existing subscribers. The surge in agentic workflow usage—especially long-running, parallelized sessions—has put unprecedented strain on our infrastructure. By stopping new enrollments, we can allocate resources more effectively and maintain service quality for current users. This pause allows us to refine our capacity management and plan for sustainable growth. Existing subscribers are not affected, and we will provide updates when sign-ups reopen. Our priority is ensuring that every active user experiences reliable performance without disruptions.

How are usage limits changing for individual plans?

We are tightening usage limits for individual plans to prevent resource overload. The Pro+ plan now offers over five times the usage limits of the Pro plan. Users on Pro who need higher capacity can upgrade to Pro+. To help you track your consumption, usage limits are now displayed directly in VS Code and the Copilot CLI. These limits are based on two factors: token consumption and a model-specific multiplier. Two types of limits exist: session limits (to prevent overloading during peak times, typically not affecting most users) and weekly limits (a cap on total weekly tokens to control costs from long-running, parallel requests). If you approach a limit, you’ll see warnings so you can adjust usage. Most users should not be impacted by these tighter limits during normal use.

What is changing with model availability, specifically Opus models?

Opus models are no longer available on the Pro plan. The Opus 4.7 model remains accessible on the Pro+ plan. Additionally, as announced in our changelog, Opus 4.5 and Opus 4.6 will be removed from Pro+. These changes reflect the higher computational costs of running advanced models, especially in agentic workflows. By focusing model availability on higher-tier plans, we can ensure that all users maintain a predictable, high-quality experience. If you rely on Opus models, upgrading to Pro+ is recommended. We continue to evaluate which models provide the best balance of capability and performance for our subscribers.

Why are these changes necessary?

Agentic workflows have fundamentally altered how GitHub Copilot operates. Long-running, parallelized sessions now regularly consume far more resources than the original plan structure was designed to support. Without action, service quality would degrade for everyone due to overuse and unpredictable demand. These changes—pausing sign-ups, tightening limits, and adjusting model availability—are essential to protect the experience of our existing customers. They allow us to stabilize costs, maintain responsiveness, and continue investing in Copilot’s development. While disruptive, they ensure that the service remains reliable and sustainable for the long term. We understand the frustration and are committed to communicating guardrails more clearly going forward.

How do the new usage limits work in practice?

GitHub Copilot now enforces two categories of usage limits: session limits and weekly limits. Session limits prevent the service from being overwhelmed during peak usage. They are set so that most users never encounter them—if you do hit a session limit, you must wait until the window resets. Weekly limits cap the total number of tokens you can consume over seven days. These were introduced to control costs from long-running, complex requests. Both limits depend on token consumption and a model multiplier. Currently, the Pro plan has lower weekly limits than Pro+. You can see your current usage directly in VS Code and Copilot CLI to avoid surprises. Most regular users should remain within these limits without disruption.

What should I do if I'm affected by these changes or want to cancel?

If you encounter unexpected usage limits or the changes do not suit your needs, you can cancel your Pro or Pro+ subscription and receive a refund for the remaining time on your current billing period. To do so, visit your Billing settings before May 20. For Pro users who need higher limits, upgrading to Pro+ is an option. We also recommend reviewing your usage patterns—if you rely heavily on agentic features, the Pro+ plan may be more appropriate. If you have further questions, support documentation is available, and we encourage feedback to help us refine these policies. We aim to make this transition as smooth as possible for all subscribers.

Will these changes impact my current experience as an existing subscriber?

As an existing subscriber, you will not lose access to your plan. However, the tighter usage limits and model changes may affect how you use Copilot—especially if you run many long agentic sessions or rely on Opus models on a Pro plan. Most users should see no immediate change in their daily workflow because the limits are set to accommodate typical usage. If you do find that your experience is hindered, you have options: upgrade to Pro+ for higher limits and continued access to Opus 4.7, or adjust your workflows to stay within the new boundaries. We are committed to maintaining a high-quality service, and these changes are designed to protect that experience for everyone.