Failing to create an ingest pipeline

Plugins Machine Learning
1

Versions (relevant - OpenSearch/Dashboard/Server OS/Browser): 2.18.0.0

Issue

I have successfully registered a huggingface model on OpenSearch, and as a next step, I went to create an ingest pipeline

    pipeline_body = {
        "processors": [
            {
                "inference": {
                    "model_id": model_id,
                    "target_field": "embedding",  # Where embeddings will be stored
                    "field_map": {
                        "ReportDescription": "text"  # Process 'ReportDescription'
                    }
                }
            }
        ]
    }
    response = client.ingest.put_pipeline(id=pipeline_id, body=pipeline_body)

And I am getting this response
RequestError(400, 'parse_exception', 'No processor type exists with name [inference]')

2

Can you share commands which are used to register/deploy models?

Before creating an ingest pipeline, ML models should be both registered and deployed to each Node(aka ML Node). Make sure the models are successfully deployed. You can check it on MachineLearning page in your Dashboards