search

Azure-Samples / azure-search-openai-demo

A sample app for the Retrieval-Augmented Generation pattern running in Azure, using Azure AI Search for retrieval and Azure OpenAI large language models to power ChatGPT-style and Q&A experiences.
https://azure.microsoft.com/products/search
MIT License
5.94k stars 4.08k forks source link

Installtion Setup #307

Open bhasampa opened 1 year ago

bhasampa commented 1 year ago

Dear Team,

When I am planning to execute the setup getting the following error:

ERROR: deployment failed: failing invoking action 'provision', error deploying infrastructure: deploying to subscription:

Deployment Error Details: InvalidTemplateDeployment: The template deployment 'openai' is not valid according to the validation procedure. The tracking id is '6777928d-1567-4630-b286-8f77cbf7cacf'. See inner errors for details. InsufficientQuota: The specified capacity '120' of account deployment is bigger than available capacity '0' for UsageName 'Tokens Per Minute (thousands) - GPT-35-Turbo'.

TraceID: 09f9bd8b6cdac42e96b5848442b7a646

But here I am trying to use the existing the deployed gpt & chat model in spite of why it is going to create new deployment model?

AZURE_OPENAI_CHATGPT_DEPLOYMENT="davinci" AZURE_OPENAI_GPT_DEPLOYMENT="chat" AZURE_OPENAI_RESOURCE_GROUP="XXXX" AZURE_OPENAI_SERVICE="XXXX"

am I missing anything can you please help on the same

Thanks, Bhaskar

clojosh commented 1 year ago

I am getting the same error and I was able to successfully deploy last Friday.

GregLin0420 commented 1 year ago

The same here.

paprocki-r commented 1 year ago

same

silvio-machado commented 1 year ago

same

tickx-cegeka commented 1 year ago

I have the same issue since this week indeed. If you need 120 for example and it is currently taking 120/120 , it still demands another 120 units to be free, although no extra units would be needed in practice. Temporary fix for us is to go to quotas and set all current deployments to a very low number such that you have enough free space again. You'd have to do this for every deployment, so I am also waiting for a fix.

bhasampa commented 1 year ago

Thanks for advise I had updated the quota in model the deployment is going fine.. but eventually we need to wait for fix form MSFT.

paprocki-r commented 1 year ago

so, temporary solution is to go to Azure AI studio, Deployments, and for each Edit deployment, Advanced, set "Tokens per Minute Rate Limit" from 120 to e.g. 2. Then deploy your accelerator

jjczopek commented 1 year ago

This is very inconvenient - having to change quotas manually, especially if the infra is deployed frequently. I hope it's gonna get fixed soon.

TimoRiikonenIF commented 1 year ago

How do you decrease the quota for the service you are deploying? Let's presume that you are using 120/120. You decrease it to 1/120. Then it tries to deploy another 120 and complains that it is more than the free quota which is 119.

jongio commented 1 year ago

This https://github.com/Azure-Samples/azure-search-openai-demo/commit/f5a7095e44bd45e224c3200874be36e93e4e0579 should help. Can you either get that patch into your current code or start from scratch and try again?

github-actions[bot] commented 1 year ago

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.