Find answers from the community

Updated 5 months ago

Amazon.titan-embed-text-v1 model identifier is invalid error

At a glance

The community member is trying to create embeddings using the amazon.titan-embed-text-v1 model, but is facing a ValidationException error. Another community member shares some code using the BedrockEmbedding class from the llama_index library, but encounters a similar InvokeModel error. The issue is resolved by using the correct model_name parameter when initializing the BedrockEmbedding instance. The community members confirm that the issue is now resolved.

CCandess

Hello,
I am trying to create embeddings using amazon.titan-embed-text-v1, I have access to it I have checked it through code as well.
However creating embeddings is not happening I have exceutive permission also.

This is the error I am facing please help.
ValidationException: An error occurred (ValidationException) when calling the InvokeModel operation: The provided model identifier is invalid.

7 comments

WWhiteFang_Jr

Hi can you share the code please

CCandess

from llama_index.core import BedrockEmbedding

model= BedrockEmbedding(model='cohere.embed-english-v3')
embed = model.get_text_embedding('this is text')

The error is on last line leading to InvokeModel error

WWhiteFang_Jr

It should be model_name

Plain Text

model=BedrockEmbedding(model_name='cohere..')

WWhiteFang_Jr

Example for this embedding class has been updated yesterday only

CCandess

Good catch, I am unable to do same for Titan models on the above code like amazon.titan-embed-text-v1

WWhiteFang_Jr

I hope with model_name now it is working for you

CCandess

Yes it is working now thank you for your help.

Add a reply