Important
Certain requests may experience rate limits.
- Rate limits restrict the number of requests that can be made within a specific time period.
- High demand may result in rate limiting.
Billing for premium requests will start on the following dates:
- May 5, 2025: Copilot Free, Copilot Pro, and Copilot Pro+
- May 12, 2025: Self-service (credit card) Copilot Business and Copilot Enterprise
- May 19, 2025: Sales-served (invoiced) Copilot Business and Copilot Enterprise
What is a request?
A request is any interaction where you ask Copilot to do something for you—whether it’s generating code, answering a question, or helping you through an extension. Each time you send a prompt in a chat window or trigger a response from Copilot, you’re making a request.
If you have Copilot Free enabled, your GitHub account comes with up to 2,000 code completions and up to 50 chats or premium requests per month.
If you're on a paid plan, you get unlimited code completions, unlimited agent requests, and unlimited chat interactions using the base model. Rate limiting is in place to accommodate for high demand. Learn more about current models and usage.
Paid plans also receive a monthly allowance of premium requests, which can be used for advanced chat interactions, code completions using premium models, and other premium features. For an overview of the amount of premium requests included in each plan, see Plans for GitHub Copilot.
Premium requests
Some Copilot features use more advanced processing power and count as premium requests. The number of premium requests a feature consumes can vary depending on the feature and the AI model used.
Premium features
The following Copilot features can use premium requests:
- Copilot Chat
- Copilot agent mode
- Copilot code review
- Copilot Extensions
Model multipliers
Each model has a premium request multiplier, based on its complexity and resource usage. Your premium request allowance is deducted according to this multiplier.
| Model | Premium requests |
|---|---|
| Base model (currently GPT-4o) 1 | 0 (paid users), 1 (Copilot Free) |
| Claude 3.5 Sonnet | 1 |
| Claude 3.7 Sonnet | 1 |
| Claude 3.7 Sonnet Thinking | 1.25 |
| Gemini 2.0 Flash | 0.25 |
| Gemini 2.5 Pro | 1 |
| GPT-4.5 | 50 |
| o1 | 10 |
| o3-mini | 0.33 |
Additional premium requests
Note
The option to purchase additional premium requests is not available to:
- Users on Copilot Free. To access more premium requests, upgrade to a paid plan.
- Users who subscribe, or have subscribed, to Copilot Pro or Copilot Pro+ through GitHub Mobile on iOS or Android.
If you use all of your premium requests, you can still use Copilot with the base model for the rest of the month. If you need more premium requests, you can upgrade to a higher plan or purchase additional premium requests. Additional premium requests beyond your plan’s included amount are billed at $0.04 USD per request.
Important
You will be able to enable additional premium requests in your account settings starting on:
- May 5, 2025: Copilot Free, Copilot Pro, and Copilot Pro+
To purchase additional premium requests, you’ll need to enable additional premium requests in your account settings first. See Managing Copilot policies as an individual subscriber or Managing policies for Copilot in your organization.
Additionally, you must set a budget in your account settings.
Example of premium request usage
Premium request usage is based on the model’s multiplier and the feature you’re using. For example:
- If you use GPT-4.5 (50× multiplier) to ask a single question in Copilot Chat, that interaction counts as 50 premium requests.
- If you're on Copilot Free, even interactions with the base model use 1 premium request each.
- If you're on a paid plan, using the base model does not count against your monthly premium request allowance.
If you've enabled additional usage, premium requests beyond your included monthly amount will be billed at $0.04 USD each.
Footnotes
-
The base model at the time of writing is GPT-4o. This is subject to change. Response times for the base model may vary during periods of high usage. Requests to the base model may be subject to rate limiting. ↩