resp = client.inference.put_model(
task_type="text_embedding",
inference_id="my-msmarco-minilm-model",
body={
"service": "elasticsearch",
"service_settings": {
"num_allocations": 1,
"num_threads": 1,
"model_id": "msmarco-MiniLM-L12-cos-v5",
},
},
)
print(resp)