An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
Hi @Swapnil Wani (Tata Consultancy Services Limited)
Answer was hinted in above comment and shared with you internally earlier for reference.
Re-emphasizing again for users restricted to use managed identity only.
As AI search is supported through private endpoint only. Using Public IPs won't work even with managed VNET.
We should create AI search resource in same region as of foundry and VNET and connect them through private endpoints.
Once they are secured with VNET, you can use a VM's Bastion browser (internal subscription has restriction)/VPN Gateway (p2s works/s2s is also supported) in same VNET.
- Please leverage Template 19 to create Foundry and AI search resource for Own VNET and tools in VNET scenario https://github.com/microsoft-foundry/foundry-samples/tree/main/infrastructure/infrastructure-setup-bicep/19-private-network-agent-toolsWorked for one of my recent interactions and saves manual network configuration. Please check out P2S/S2S documentation with networking team to configure if accessing with Bastion browser does not work
- Please leverage Template 18 for managed VNET (Bug fixed on using own vnet recently)
(Bring your own VNET scenario/ Managed VNET along own VNET)
Thank you for understanding our constraints on helping on networking part.