inference-apis.asciidoc 2.2 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253
  1. [role="xpack"]
  2. [[inference-apis]]
  3. == {infer-cap} APIs
  4. experimental[]
  5. IMPORTANT: The {infer} APIs enable you to use certain services, such as built-in
  6. {ml} models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Azure,
  7. Google AI Studio or Hugging Face. For built-in models and models uploaded
  8. through Eland, the {infer} APIs offer an alternative way to use and manage
  9. trained models. However, if you do not plan to use the {infer} APIs to use these
  10. models or if you want to use non-NLP models, use the
  11. <<ml-df-trained-models-apis>>.
  12. The {infer} APIs enable you to create {infer} endpoints and use {ml} models of
  13. different providers - such as Amazon Bedrock, Anthropic, Azure AI Studio,
  14. Cohere, Google AI, Mistral, OpenAI, or HuggingFace - as a service. Use
  15. the following APIs to manage {infer} models and perform {infer}:
  16. * <<delete-inference-api>>
  17. * <<get-inference-api>>
  18. * <<post-inference-api>>
  19. * <<put-inference-api>>
  20. [[inference-landscape]]
  21. .A representation of the Elastic inference landscape
  22. image::images/inference-landscape.png[A representation of the Elastic inference landscape,align="center"]
  23. An {infer} endpoint enables you to use the corresponding {ml} model without
  24. manual deployment and apply it to your data at ingestion time through
  25. <<semantic-search-semantic-text, semantic text>>.
  26. Choose a model from your provider or use ELSER – a retrieval model trained by
  27. Elastic –, then create an {infer} endpoint by the <<put-inference-api>>.
  28. Now use <<semantic-search-semantic-text, semantic text>> to perform
  29. <<semantic-search, semantic search>> on your data.
  30. include::delete-inference.asciidoc[]
  31. include::get-inference.asciidoc[]
  32. include::post-inference.asciidoc[]
  33. include::put-inference.asciidoc[]
  34. include::service-amazon-bedrock.asciidoc[]
  35. include::service-anthropic.asciidoc[]
  36. include::service-azure-ai-studio.asciidoc[]
  37. include::service-azure-openai.asciidoc[]
  38. include::service-cohere.asciidoc[]
  39. include::service-elasticsearch.asciidoc[]
  40. include::service-elser.asciidoc[]
  41. include::service-google-ai-studio.asciidoc[]
  42. include::service-google-vertex-ai.asciidoc[]
  43. include::service-hugging-face.asciidoc[]
  44. include::service-mistral.asciidoc[]
  45. include::service-openai.asciidoc[]