Availability of models for deployments by location

Question

Hi

I live in the UK and have written a webapp with both GPT-3 and GPT-4 model deployments.

I use Digital Ocean to host my app and am finding that my GPT-3 deployment works although the GPT-4 doesn't even though both work locally.

I have no idea of the location of Digital Oceans servers are and am wondering if their location is making a difference.

What determines the location of a deployment and thus if a model is available?

Many thanks!

Justin

Answer 1

Justin Baron Greetings!

Can you share more details on what is the exact issue with GPT-4?

Yes, the location of the servers where your web app is hosted can make a difference in the availability of your GPT-4 deployment.

What determines the location of a deployment and thus if a model is available?

It depends on the specific requirements. The location determines the region where the deployment is created and where the model is available. If you want to use a model in a different region, you need to create a new deployment in that region.

Azure OpenAI Service is provides diverse set of models with different capabilities and price points. Model availability varies by region.

Do let me know if you have any further queries.