C1M5 Error code: 400 - {"message": "Unable to access non-serverless model meta-llama/Llama-3.2-3B-Instruct-Turbo

I am getting the following error when I am trying the graded assignment in Module 5. I use DeepLearning.AI Error code: 400 - {“message”: “Unable to access non-serverless model meta-llama/Llama-3.2-3B-Instruct-Turbo. Please visit to create and start a new dedicated endpoint for the model.”, “type_”: “invalid_request_error”, “code”: “model_not_available”}

hi @kalyansri2508

All labs of RAG course we’re recently updated for this issue. So one needs to click on 3 dots on right upper corner, then click Restore Original version to have the updated notebook in your work environment.

Let me know if issue still persist.

Regards

Dr. Deepti

hi @Deepti_Prasad I read the forum and tried Restore Original version already. The issue still persists specifically with check_if_faq_or_product(query, simplified = True)

1 Like

@kalyansri2508

I have already reported this issue to staff yesterday but as it’s weekend, staff will probably look on this on Monday.

The issue persists because the function recall are still invoking codes based on old deprecated model as per my understanding after reviewing the lab’s metadata.

regards

Dr. Deepti

2 Likes

@kalyansri2508

staff update

together.ai has deprecated a lot of models in the past days resulting in concurrent issues with labs of RAG courses.

Lab assignment 3 to 5 and all ungraded labs are updated now.

You need to Restore original version, to get the updated notebook.

I also would recommend to first delete the old file, and restore original version, so any cache issue is not encountered when running codes again.

Courses labs are created with multiple channels, so labs gets affected when any changes.made at one end and can take multiple attempts to resolve issues.

Let us know if you still continue to have issue

Regards

Dr. Deepti

1 Like

@Deepti_Prasad Thank you. It is working now

1 Like

I’m still having the problem. I deleted all files, then used the “Restore original version” command. I’m using deeplearning.ai. Here’s the error I got in the Phoenix trace:

exceptionError code: 400 - {“message”: “Unable to access non-serverless model meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo. Please visit Together AI | The AI Native Cloud to create and start a new dedicated endpoint for the model.”, “type_”: “invalid_request_error”, “code”: “model_not_available”}

1 Like

please post screenshot for which code did you encounter this error. Also as a safeguard I also tell learners to delete their cache and browsing history before they try to get the updated notebook

I just did delete my cache and browsing history, and I’m still getting the error. Here are the screen shots.

lab assignment number?

Also based on the error, I still feel you are not working on updated notebook.

So please make sure you have clicked on the 3 dots on right top corner and then clicked Restore original.

let me know which lab assignment number it is, so I can check in my lab environment

C1M5. I did click the 3 dots last night after deleting all the files.

1 Like

ok let me check.

I am volunteer mentor for the course, and have updated the staff about this. but unfortunately as it is weekend, they will probably check when weekday starts in their timeline.

until then me as a mentor and you as learner can only try to find solution from our side.

Apologies for these setback as labs of RAG have been encountering such issues, staff has been consistently updating to resolve issue, yet newer issues are probably arising.

Regards
Dr. Deepti

1 Like

hi @christophomos

I just check the lab, looks like a new issue with json load is using together api key, and it is throwing exception when recalling the meta llama. We will have to wait for @lucas.coutinho to resolve this issue.

Regards
Dr. Deepti

I appreciate your help, Dr. Deepti!

1 Like

Hi all.

We had another model deprecated by together.ai, I’m right now working in fixing it, replacing by another model. It might take a few hours. I will let you all know once it is fixed.

Cheers,
Lucas

3 Likes

@christophomos

please update with a response if your issue is resolved as the staff has addressed your issue.

Regards
Dr. Deepti

1 Like

The assignment now works perfectly. I have now passed the course. Thanks @Deepti_Prasad and @lucas.coutinho

1 Like