capabilities.work

Create an Elasticsearch-hosted model inference endpoint

One step with Elastic Stack. We can help you get access.

Get something done

Create an Elasticsearch-hosted model inference endpoint

with Elastic Stack · Artificial Intelligence Management

Use Elastic Stack: its Elasticsearch Request & Response Specification Inference API lets you create an Elasticsearch-hosted model inference endpoint in a single step (PUT /_inference/{task_type}/{elasticsearch_inference_id}). capabilities.work can help you get signed up and connected.

  • How do I deploy a model running inside my own cluster, like E5 or one uploaded with Eland, as an inference endpoint?
  • Do I need to create an endpoint for ELSER or E5 if the deployment already has preconfigured ones?
Say it to your AI assistant
  • Create an elasticsearch-service {task_type} endpoint {elasticsearch_inference_id} for a model in my cluster.
  • Deploy the in-cluster model as {elasticsearch_inference_id} for {task_type} with {service_settings}.
For your developer

PUT /_inference/{task_type}/{elasticsearch_inference_id} in Elasticsearch Request & Response Specification Inference API. Full spec on APIs.io.

More you can do with Elastic Stack

Get ML memory usage across all nodesGet ML memory usage for one nodeGet machine learning defaults and limitsToggle ML upgrade modeSync ML saved objects in the default spaceChange which spaces ML jobs belong toChange which spaces trained models belong toClear a trained model deployment's inference cacheGet trained model configs by IDUpload a trained modelDelete an unreferenced trained modelCreate or reassign a trained model alias