> For the complete documentation index, see [llms.txt](https://docs.groundcover.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.groundcover.com/architecture/byoc/setup-byoc-with-azure/enable-ai-models-for-agent-mode.md). # Enable AI models for Agent Mode Agent Mode on Azure BYOC runs against **Anthropic Claude** models hosted on **Microsoft Foundry**. There are no manual setup steps for you: the groundcover team provisions everything below in your Azure subscription during BYOC setup. ## What groundcover provisions | Resource | Name / Identifier | Location | Purpose | | ------------------------------------- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Resource group | `-foundry` | East US 2, or another [Foundry-supported region](https://learn.microsoft.com/azure/ai-foundry/) | Dedicated, separate from your cluster's resource group. Anthropic models on Foundry are only available in select regions, so Agent Mode infrastructure may live cross-region from your AKS cluster. | | User-Assigned Managed Identity (UAMI) | `agent-service-` | Your cluster's resource group | Identity the agent-service pod uses to call Foundry. | | Federated identity credential | On the UAMI above | n/a | Scoped to the `agent-service` Kubernetes ServiceAccount. Lets the pod authenticate to Foundry via AKS workload identity, with no API key. | | Microsoft Foundry account | `gc-` | Same RG as above | A Cognitive Services resource of kind `AIServices`. This is the endpoint your pod calls for Claude inference. | | Foundry project | Inside the Foundry account | n/a | Required organizational unit for Foundry deployments. | | Model deployment | `claude-sonnet-4-6` (50 TPM capacity) | n/a | Anthropic Claude Sonnet 4.6. | | Model deployment | `claude-haiku-4-5` (50 TPM capacity) | n/a | Anthropic Claude Haiku 4.5. | | Role assignment | **Azure AI User** on the Foundry account, granted to the `agent-service` UAMI | n/a | Minimum-privilege built-in role allowing `POST` to `/anthropic/v1/messages`. | | Marketplace agreement | Anthropic publisher terms on your subscription | Subscription | Required before any paid SaaS model can be deployed. | | Resource provider registration | `Microsoft.SaaS` | Subscription | Idempotent. No effect if already registered. | ## Cost You pay only for **actual model inference**. All the infrastructure above (the Foundry account, project, model deployments, UAMI, resource group, role assignments, and marketplace agreement) is free. Anthropic model inference is billed pay-as-you-go on tokens. Microsoft Foundry uses **Anthropic's standard API pricing** for Claude models on the Marketplace. Approximate rates at the time of writing: | Model | Input (per M tokens) | Output (per M tokens) | | ----------------- | -------------------- | --------------------- | | Claude Sonnet 4.6 | \~$3 | \~$15 | | Claude Haiku 4.5 | \~$1 | \~$5 | Rates may change. For current pricing, see the [Azure AI Foundry pricing page](https://azure.microsoft.com/pricing/details/ai-foundry-models/). To cap spending across your organization or specific users, see [Cost Management](/use-groundcover/agent-mode/cost-management.md). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.groundcover.com/architecture/byoc/setup-byoc-with-azure/enable-ai-models-for-agent-mode.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.