Open bmis opened 10 months ago
Not yet. Meanwhile, one idea would be to have APIM fronting this load balancer and use its logging capabilities. There is one Azure reference architecture for this: https://learn.microsoft.com/en-us/azure/architecture/ai-ml/openai/architecture/log-monitor-azure-openai
for a logging example (in this case token usage) have a look at this repo https://github.com/Azure/enterprise-azureai
@andredewes : I was also on the path of using YARP for monitoring and took the liberty to use your priority-loadbalancing code. Thanks for the great work!
Does this solution have the ability to log prompts going to Azure Open AI and responses from Azure Open AI? For example using Azure Monitor?
Thanks.