|
| 1 | +--- |
| 2 | +title: 'GitHub Copilot usage and models' |
| 3 | +description: Learn how to manage your GitHub Copilot usage and models directly within Visual Studio. |
| 4 | +ms.date: 6/4/2025 |
| 5 | +ms.topic: overview |
| 6 | +author: anandmeg |
| 7 | +ms.author: meghaanand |
| 8 | +ms.manager: mijacobs |
| 9 | +ms.subservice: ai-tools |
| 10 | +ms.collection: ce-skilling-ai-copilot |
| 11 | +monikerRange: '>= vs-2022' |
| 12 | +--- |
| 13 | +# Copilot usage and models in Visual Studio |
| 14 | + |
| 15 | +GitHub Copilot now follows a [new billing model](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests). With this update that impacts usage and billing for [premium requests](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests#premium-requests), Visual Studio has introduced new features to help you better manage your Copilot usage and the models you work with. |
| 16 | + |
| 17 | +## What's changed? |
| 18 | + |
| 19 | +If you’re a **paid Copilot user**, your plan now includes a **[monthly allowance of premium requests](https://docs.github.com/en/copilot/about-github-copilot/plans-for-github-copilot#comparing-copilot-plans)**. These requests are used whenever you interact with Copilot using premium models, like Claude Sonnet 3.7. Once you reach your quota, Copilot in Visual Studio automatically switches to a base model, ensuring uninterrupted workflow. |
| 20 | + |
| 21 | +For users on **Copilot Free**, your account comes with up to 2,000 code completions and up to 50 chats or premium requests per month. |
| 22 | + |
| 23 | +## What’s new in Visual Studio? |
| 24 | + |
| 25 | +New tools and smarter defaults make it easier to track usage and get the most out of Copilot. |
| 26 | + |
| 27 | +### Smarter default models |
| 28 | + |
| 29 | +Visual Studio now defaults to **GPT-4.1**, a fast and balanced model. If you exhaust your premium quota, it automatically switches to the **fallback model**, ensuring uninterrupted access to powerful features at no extra cost. |
| 30 | + |
| 31 | +> 💡 Once you select a model, it stays active for your current chat session—no need to reselect it between threads. |
| 32 | +
|
| 33 | +### Smarter usage insights |
| 34 | + |
| 35 | +To track your Copilot usage: |
| 36 | + |
| 37 | +1. Select the **Copilot badge** in the top-right corner of Visual Studio. |
| 38 | +2. Select **Copilot Consumptions**. |
| 39 | + |
| 40 | +:::image type="content" source="media/vs-2022/copilot-usage-models/copilot-consumptions-panel.png" alt-text="Screenshot that shows Copilot Consumptions panel." lightbox="media/vs-2022/copilot-usage-models/copilot-consumptions-panel.png"::: |
| 41 | + |
| 42 | +Here, you can view your usage across chat, auto (agent) mode, and other features. You can also select **Get more requests** to update your subscription or adjust Overage settings directly on *GitHub.com*. |
| 43 | + |
| 44 | +### Smarter fallback behavior |
| 45 | + |
| 46 | +- **Overages ON**: You can keep using premium models beyond your monthly quota, with any extra usage billed to your account. |
| 47 | +- **Overages OFF**: Copilot automatically switches to the fallback model once you reach your quota, preventing unexpected charges. |
| 48 | + |
| 49 | +To control model usage in non-chat interactions, configure the setting **Enhance non-chat requests with premium models** in **GitHub** > **Copilot** > **Editor**. |
| 50 | + |
| 51 | +### Easier model management |
| 52 | + |
| 53 | +Use the updated **model picker** experience. |
| 54 | + |
| 55 | +> If a model is supported by your plan but not yet enabled, Visual Studio prompts you to activate it—no need to leave the IDE. |
| 56 | +
|
| 57 | +### Models available |
| 58 | + |
| 59 | +- Gemini 2.5 Pro |
| 60 | +- Gemini 2.0 Flash |
| 61 | +- Claude Sonnet 3.7 |
| 62 | +- Claude Sonnet 3.5 |
| 63 | +- GPT-4.1 |
| 64 | +- GPT-4.5 |
| 65 | +- OpenAI o3 Mini |
| 66 | +- GPT-4o |
| 67 | + |
| 68 | +> ℹ️ Each model has a premium request multiplier, for example, GPT-4.5 counts as 50 premium requests per use. You’ll see the exact multiplier in the **model picker** before selecting. |
| 69 | +
|
| 70 | +:::image type="content" source="media/vs-2022/copilot-usage-models/model-picker.png" alt-text="Screenshot that shows Pick Model dropdown." lightbox="media/vs-2022/copilot-usage-models/model-picker.png"::: |
| 71 | + |
| 72 | +## Frequently asked questions |
| 73 | + |
| 74 | +#### Do Completions count as premium requests? |
| 75 | + |
| 76 | +**Code completions** do *not* count against your premium request balance for paid users. |
| 77 | + |
| 78 | +#### What are premium requests? |
| 79 | + |
| 80 | +Some Copilot features use more advanced processing power and count as premium requests. See [premium requests and premium features](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests#premium-requests) to learn more. |
| 81 | + |
| 82 | +#### Who can purchase additional premium requests? |
| 83 | + |
| 84 | +To learn more, see [Additional premium requests](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests#additional-premium-requests). |
| 85 | + |
| 86 | +#### How is premium request usage assessed? |
| 87 | + |
| 88 | +Premium request usage is based on the [model’s multiplier](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests#model-multipliers) and the feature you’re using. To learn more, see [Example of premium request usage](https://docs.github.com/en/copilot/managing-copilot/monitoring-usage-and-entitlements/about-premium-requests#example-of-premium-request-usage.) |
| 89 | + |
| 90 | +## Related content |
| 91 | + |
| 92 | +- [Install and manage GitHub Copilot in Visual Studio](visual-studio-github-copilot-install-and-states.md) |
| 93 | +- [About GitHub Copilot Free](https://aka.ms/ghdocscopilotfreepage) |
0 commit comments