# Managing models > For the complete documentation index, see [llms.txt](https://www.jetbrains.com.cn/en-us/help/jetbrains-console/main/llms.txt). You can add, configure, and remove models to control which of them are available through each AI provider in your organization. > **Note:** > For the JetBrains AI provider, model management is not allowed. You cannot add, configure, or delete any models for this provider. Each provider card displays models that users and service accounts in your organization can access. For each model, the following information is available: * The model name shown in JetBrains AI tools, such as [AI Assistant](https://www.jetbrains.com.cn/en-us/help/ai-assistant/about-ai-assistant.html) and [JetBrains Air](https://www.jetbrains.com.cn/en-us/help/air/getting-started.html). * The model ID used by the provider. * The model status, if any: * Deprecated: the provider has deprecated the model. * Error: the latest [connectivity check](#check-model-connectivity) for the model failed. * No access: you have no [AI access](ai-access.html), so you cannot see the connection status of the model. * The JetBrains-specific model ID used by JetBrains to uniquely identify the model. This value may differ from the provider's model ID. To view the JetBrains-specific model ID, hover over the model name in the Models table. > **Note:** > Viewing the page requires the [View AI settings](roles.html#permissions_view_ai_settings) permission. Managing provider models requires the [Manage AI settings](roles.html#permissions_manage_ai_settings) or [Manage organization](roles.html#permissions_manage_org) permission. Procedure: Add a model 1. In the sidebar, under AI governance, select Providers. If you don't see this page, or the controls on this page are inactive, your role doesn't have the necessary [permissions](roles.html#permissions). 2. In the provider card, open the Add model dropdown and select a model. The dropdown displays models supported by the provider. 3. In the Provider model ID field, review and adjust the pre-filled model ID. It must be the exact model ID that the provider uses for the model in your region. Otherwise, the [connectivity check](#check-model-connectivity) might fail. > **Tip:** > For Amazon Bedrock, you must specify the exact ID that Amazon Bedrock uses for the model in your region. To learn how to get the correct ID, refer to [Amazon Bedrock model ID](ai-providers-amazon-bedrock.html#amazon-bedrock-model-id). 4. Click Save. JetBrains Central Console first adds the model and then automatically checks if it is reachable and responds to requests. If the model cannot be added, nothing is created: the dialog stays open and shows an error message. If the model is added, it appears in the Models table regardless of the check result. If the check fails, the Error status appears in the corresponding table row. > **Tip:** > If the check fails, ensure the configuration is correct and the model is available on the provider's side. Procedure: Edit a model configuration You can edit a model configuration by modifying its Provider model ID. For example, you may need to specify a different model version or a region-specific ID supported by the provider. 1. In the sidebar, under AI governance, select Providers. If you don't see this page, or the controls on this page are inactive, your role doesn't have the necessary [permissions](roles.html#permissions). 2. In the provider card, click ![Ellipsis icon](images/union.svg) in the corresponding row of the Models table, and select Edit model settings. 3. In the Provider model ID field, specify the exact model ID that the provider uses for the model in your region. Otherwise, the [connectivity check](#check-model-connectivity) might fail. > **Tip:** > For Amazon Bedrock, you must specify the exact ID that Amazon Bedrock uses for the model in your region. To learn how to get the correct ID, refer to [Amazon Bedrock model ID](ai-providers-amazon-bedrock.html#amazon-bedrock-model-id). 4. Click Save. JetBrains Central Console first saves the changes and then automatically checks if the model is reachable and responds to requests. If the changes cannot be saved, the configuration stays as it was: the dialog stays open and shows an error message. If the changes are saved, the provider model ID is updated in the Models table regardless of the check result. If the check fails, the Error status appears in the corresponding table row. > **Tip:** > If the check fails, ensure the configuration is correct and the model is available on the provider's side. Procedure: Check model connectivity You can manually check if a specific model is reachable and responds to requests. > **Note:** > Checking model connectivity requires [AI access](ai-access.html) to be granted to your account, as the check sends a request to the model on your behalf. > > > > If you have the [Manage AI access](roles.html#permissions_manage_ai_access) or [Manage organization](roles.html#permissions_manage_org) permission, open the [AI access](ai-access.html) page and grant yourself AI access. Otherwise, ask an administrator to grant it to you. Checking model connectivity sends a real request to the model through your provider: a short prompt with a limit of 500 output tokens. Your JetBrains AI credits are not spent on this request, but your provider may bill you for it. 1. In the sidebar, under AI governance, select Providers. If you don't see this page, or the controls on this page are inactive, your role doesn't have the necessary [permissions](roles.html#permissions). 2. In the provider card, click ![Ellipsis icon](images/union.svg) in the corresponding row of the Models table, and select Check connection. If the check passes, you see a success message in the lower-right corner stating that the model is reachable and responds to requests. The Error status from the previous check, if any, disappears. If the check fails, the Error status appears in the corresponding table row. Hover over the status to see the error details. If AI access is not granted to you, the No access status appears in the row instead of the check result. In this case, you cannot see the connection status of the model at all, even if the model has an error. The Error status is not stored: it disappears when you reload the page, leave it and come back, or when you edit or save the provider configuration. It appears again only after the next check. > **Note:** > Running Check connection will not clear the model's Deprecated status assigned by the provider. Procedure: Remove a model 1. In the sidebar, under AI governance, select Providers. If you don't see this page, or the controls on this page are inactive, your role doesn't have the necessary [permissions](roles.html#permissions). 2. In the provider card, click ![Ellipsis icon](images/union.svg) in the corresponding row of the Models table, and select Remove model. 3. In the dialog, click Remove to confirm the action. The model is removed from the table. Users and service accounts who had access to this model can no longer use it.