where did gpt-3.5-turbo-0613 and gpt-3.5-turbo-16k go?

Question

where did gpt-3.5-turbo-0613 and gpt-3.5-turbo-16k go?

Jerzy Czopek 30

So I have been able to deploy both gpt-35-turbo 0613 and gpt-35-turbo-16k recently, but now I can't do that. I can't change version of the deployed model, and can't find 16k base model at all.

This is what I see in Azure portal:
User's image

What happened and when the models will become avaialble again?

zm zhang 0 Reputation points

2023-07-13T00:48:24.59+00:00

the same problem
Jerzy Czopek 30 Reputation points

2023-07-13T06:48:55.87+00:00

@AshokPeddakotla-MSFT please share updates once resolved. Is there a place where we can look this things by our own? blog or something?
AshokPeddakotla-MSFT 35,976 Reputation points Moderator

2023-07-13T15:33:41.23+00:00

Jerzy Czopek I will update here once I hear anything about the availability. Since this is internal I'm not really sure if you can check anything on the blogs. If team adds any posts I will add as well.
Raman Jhajj 15 Reputation points

2023-07-17T19:09:46.02+00:00

@AshokPeddakotla-MSFT Is there any update on this? We are still unable to deploy 16K model in any region.
Chris Rickard 5 Reputation points

2023-07-18T01:50:14.83+00:00

Very inconsistent info going around.. it looks like we should have it in East US, but not there.
Richard Murray-Clark 0 Reputation points

2023-07-19T09:19:13.7033333+00:00

@AshokPeddakotla-MSFT Without the 0613 update, we are unable to get consistent outputs that would be fixed by the function calling feature. When are you planning on releasing the new model?
zm zhang 0 Reputation points

2023-07-19T09:43:22.6966667+00:00

it is available now in East US
AshokPeddakotla-MSFT 35,976 Reputation points Moderator

2023-07-20T09:52:00.11+00:00

Richard Murray-Clark & zm zhang At this time these models are disabled in most regions. Our team is actively working on deploying fix for some issues we found with these models, this should be resolved by today or tomorrow. Please stay tuned for further updates. I will update here once I get information from the team.
Appreciate your time and patience.

1 answer

Your answer

zm zhang 0 Reputation points

2023-07-13T00:48:24.59+00:00

the same problem
Jerzy Czopek 30 Reputation points

2023-07-13T06:48:55.87+00:00

@AshokPeddakotla-MSFT please share updates once resolved. Is there a place where we can look this things by our own? blog or something?
AshokPeddakotla-MSFT 35,976 Reputation points Moderator

2023-07-13T15:33:41.23+00:00

Jerzy Czopek I will update here once I hear anything about the availability. Since this is internal I'm not really sure if you can check anything on the blogs. If team adds any posts I will add as well.
Raman Jhajj 15 Reputation points

2023-07-17T19:09:46.02+00:00

@AshokPeddakotla-MSFT Is there any update on this? We are still unable to deploy 16K model in any region.
Chris Rickard 5 Reputation points

2023-07-18T01:50:14.83+00:00

Very inconsistent info going around.. it looks like we should have it in East US, but not there.
Richard Murray-Clark 0 Reputation points

2023-07-19T09:19:13.7033333+00:00

@AshokPeddakotla-MSFT Without the 0613 update, we are unable to get consistent outputs that would be fixed by the function calling feature. When are you planning on releasing the new model?
zm zhang 0 Reputation points

2023-07-19T09:43:22.6966667+00:00

it is available now in East US
AshokPeddakotla-MSFT 35,976 Reputation points Moderator

2023-07-20T09:52:00.11+00:00

Richard Murray-Clark & zm zhang At this time these models are disabled in most regions. Our team is actively working on deploying fix for some issues we found with these models, this should be resolved by today or tomorrow. Please stay tuned for further updates. I will update here once I get information from the team.
Appreciate your time and patience.

Answer 1

Jerzy Czopek & zm zhang Greetings!

So I have been able to deploy both gpt-35-turbo 0613 and gpt-35-turbo-16k recently, but now I can't do that. I can't change version of the deployed model, and can't find 16k base model at all.

I have checked internally on this ask and below is an update.

In East US, 0613 model has been temporarily disabled due to the current investigations of customer reported issues. We expect this should be resolved in the next week.

Jerzy Czopek, zm zhang , Raman Jhajj , Chris Rickard & Richard Murray-Clark

Update: Good news!

gpt-35-turbo 0613 & gpt-35-turbo-16k are available now in East US for deployment. Also, the limit related to 8k is resolved as well. Let us know if you are still seeing the errors.

I just tried and able to deploy successfully.

User's image

If the response helped, please do click Accept Answer and Yes for was this answer helpful.

Doing so would help other community members with similar issue identify the solution. I highly appreciate your contribution to the community.

Deleted

This comment has been deleted due to a violation of our Code of Conduct. The comment was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

Share via

where did gpt-3.5-turbo-0613 and gpt-3.5-turbo-16k go?

1 answer

Your answer