How to connect Qlik to Meta's Llama model ?

[Kartik2]

I was trying to integrate Qlik with Llama ,I was using the default Meta(Amazon_Bedrock) anaytics connection in Qlik cloud.

The connection was created successfully and i used the below script for testing

QuestionsTable:

Load * Inline [

Question_ID,Question

1,"show me general ways to help boost a company's sales"
2,"what are the core concepts in accounting?"
3,"how large is the sun?"
];

[Meta]:
LOAD
[generation],
[prompt_token_count],
[generation_token_count],
[stop_reason],
[RowId]
EXTENSION endpoints.ScriptEval('{"RequestType":"endpoint", "endpoint":{"connectionname":"Meta_(Amazon_Bedrock)"}}', QuestionsTable{Question_ID,Question});

exit script


But i got an ERROR  

"  Connector reply error: grpc::StatusCode::FAILED_PRECONDITION: 'Error while converting required format. Status(StatusCode="FailedPrecondition", Detail="Error returned by endpoint: NotFound Could not resolve the foundation model from the provided model identifier.")'   " 

Any Solutions ??

[jprdonnelly]

@Kartik2 - if you are trying to use a model from Meta other than:

• Llama 2 Chat 13B

• Llama 2 Chat 70B

the built-in Bedrock Meta connector will not work. If attempting to utilize Llama 3 or another model, you can use the default REST connector, though it may require additional parameters (the REST interactivity wouldn't be simplified in this case, as it is with the branded connectors and the model supported from each).

In that scenario, using the default REST connector, you can leverage any LLM API you want, including custom ones. The API endpoint just needs to be available to your Qlik Cloud tenant (and this cannot be through the Data Gateway - Direct Access).

Qlik has a lot of improvements in this area just over the horizon, so please stay tuned for Qlik Connect to see some of the innovations we are bringing to this space!