diff --git a/output/openapi/elasticsearch-openapi.json b/output/openapi/elasticsearch-openapi.json index ed76c4dd31..9a02079011 100644 --- a/output/openapi/elasticsearch-openapi.json +++ b/output/openapi/elasticsearch-openapi.json @@ -20573,7 +20573,7 @@ "inference" ], "summary": "Create an inference endpoint", - "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", + "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", "operationId": "inference-put", "parameters": [ { @@ -20694,7 +20694,7 @@ "inference" ], "summary": "Create an inference endpoint", - "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", + "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", "operationId": "inference-put-1", "parameters": [ { @@ -21198,6 +21198,11 @@ "summary": "A completion task", "description": "Run `PUT _inference/completion/azure_ai_studio_completion` to create an inference endpoint that performs a completion task.", "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"databricks\",\n \"endpoint_type\": \"realtime\"\n }\n}" + }, + "PutAzureAiStudioRequestExample3": { + "summary": "A rerank task", + "description": "Run `PUT _inference/rerank/azure_ai_studio_rerank` to create an inference endpoint that performs a rerank task.", + "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"cohere\",\n \"endpoint_type\": \"token\"\n }\n}" } } } @@ -88949,6 +88954,7 @@ "type": "string", "enum": [ "completion", + "rerank", "text_embedding" ] }, @@ -89017,6 +89023,14 @@ "user": { "description": "For a `text_embedding` task, specify the user issuing the request.\nThis information can be used for abuse detection.", "type": "string" + }, + "return_documents": { + "description": "For a `rerank` task, return doc text within the results.", + "type": "boolean" + }, + "top_n": { + "description": "For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.", + "type": "number" } } }, @@ -89047,7 +89061,8 @@ "type": "string", "enum": [ "text_embedding", - "completion" + "completion", + "rerank" ] }, "inference._types.AzureOpenAITaskType": { diff --git a/output/openapi/elasticsearch-serverless-openapi.json b/output/openapi/elasticsearch-serverless-openapi.json index a827981352..f6993c2783 100644 --- a/output/openapi/elasticsearch-serverless-openapi.json +++ b/output/openapi/elasticsearch-serverless-openapi.json @@ -11368,7 +11368,7 @@ "inference" ], "summary": "Create an inference endpoint", - "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", + "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", "operationId": "inference-put", "parameters": [ { @@ -11489,7 +11489,7 @@ "inference" ], "summary": "Create an inference endpoint", - "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", + "description": "IMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)\n\n## Required authorization\n\n* Cluster privileges: `manage_inference`\n", "operationId": "inference-put-1", "parameters": [ { @@ -11993,6 +11993,11 @@ "summary": "A completion task", "description": "Run `PUT _inference/completion/azure_ai_studio_completion` to create an inference endpoint that performs a completion task.", "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"databricks\",\n \"endpoint_type\": \"realtime\"\n }\n}" + }, + "PutAzureAiStudioRequestExample3": { + "summary": "A rerank task", + "description": "Run `PUT _inference/rerank/azure_ai_studio_rerank` to create an inference endpoint that performs a rerank task.", + "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"cohere\",\n \"endpoint_type\": \"token\"\n }\n}" } } } @@ -56271,6 +56276,7 @@ "type": "string", "enum": [ "completion", + "rerank", "text_embedding" ] }, @@ -56339,6 +56345,14 @@ "user": { "description": "For a `text_embedding` task, specify the user issuing the request.\nThis information can be used for abuse detection.", "type": "string" + }, + "return_documents": { + "description": "For a `rerank` task, return doc text within the results.", + "type": "boolean" + }, + "top_n": { + "description": "For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.", + "type": "number" } } }, @@ -56369,7 +56383,8 @@ "type": "string", "enum": [ "text_embedding", - "completion" + "completion", + "rerank" ] }, "inference._types.AzureOpenAITaskType": { diff --git a/output/schema/schema.json b/output/schema/schema.json index 4eef88269d..1a4e416297 100644 --- a/output/schema/schema.json +++ b/output/schema/schema.json @@ -9920,7 +9920,7 @@ "visibility": "public" } }, - "description": "Create an inference endpoint.\n\nIMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)", + "description": "Create an inference endpoint.\n\nIMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)", "docId": "inference-api-put", "docUrl": "https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put", "extPreviousVersionDocUrl": "https://www.elastic.co/guide/en/elasticsearch/reference/8.18/put-inference-api.html", @@ -168123,7 +168123,7 @@ "name": "AzureAiStudioServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L577-L579" + "specLocation": "inference/_types/CommonTypes.ts#L587-L589" }, { "kind": "interface", @@ -168192,9 +168192,33 @@ "namespace": "_builtins" } } + }, + { + "description": "For a `rerank` task, return doc text within the results.", + "name": "return_documents", + "required": false, + "type": { + "kind": "instance_of", + "type": { + "name": "boolean", + "namespace": "_builtins" + } + } + }, + { + "description": "For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.", + "name": "top_n", + "required": false, + "type": { + "kind": "instance_of", + "type": { + "name": "integer", + "namespace": "_types" + } + } } ], - "specLocation": "inference/_types/CommonTypes.ts#L542-L570" + "specLocation": "inference/_types/CommonTypes.ts#L542-L579" }, { "kind": "enum", @@ -168202,6 +168226,9 @@ { "name": "completion" }, + { + "name": "rerank" + }, { "name": "text_embedding" } @@ -168210,7 +168237,7 @@ "name": "AzureAiStudioTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L572-L575" + "specLocation": "inference/_types/CommonTypes.ts#L581-L585" }, { "kind": "interface", @@ -168302,7 +168329,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L581-L626" + "specLocation": "inference/_types/CommonTypes.ts#L591-L636" }, { "kind": "enum", @@ -168315,7 +168342,7 @@ "name": "AzureOpenAIServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L641-L643" + "specLocation": "inference/_types/CommonTypes.ts#L651-L653" }, { "kind": "interface", @@ -168337,7 +168364,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L628-L634" + "specLocation": "inference/_types/CommonTypes.ts#L638-L644" }, { "kind": "enum", @@ -168353,7 +168380,7 @@ "name": "AzureOpenAITaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L636-L639" + "specLocation": "inference/_types/CommonTypes.ts#L646-L649" }, { "kind": "enum", @@ -168378,7 +168405,7 @@ "name": "CohereEmbeddingType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L698-L704" + "specLocation": "inference/_types/CommonTypes.ts#L708-L714" }, { "kind": "enum", @@ -168400,7 +168427,7 @@ "name": "CohereInputType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L706-L711" + "specLocation": "inference/_types/CommonTypes.ts#L716-L721" }, { "kind": "interface", @@ -168473,7 +168500,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L645-L686" + "specLocation": "inference/_types/CommonTypes.ts#L655-L696" }, { "kind": "enum", @@ -168486,7 +168513,7 @@ "name": "CohereServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L694-L696" + "specLocation": "inference/_types/CommonTypes.ts#L704-L706" }, { "kind": "enum", @@ -168505,7 +168532,7 @@ "name": "CohereSimilarityType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L713-L717" + "specLocation": "inference/_types/CommonTypes.ts#L723-L727" }, { "kind": "interface", @@ -168563,7 +168590,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L725-L757" + "specLocation": "inference/_types/CommonTypes.ts#L735-L767" }, { "kind": "enum", @@ -168582,7 +168609,7 @@ "name": "CohereTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L688-L692" + "specLocation": "inference/_types/CommonTypes.ts#L698-L702" }, { "kind": "enum", @@ -168601,7 +168628,7 @@ "name": "CohereTruncateType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L719-L723" + "specLocation": "inference/_types/CommonTypes.ts#L729-L733" }, { "kind": "interface", @@ -168884,7 +168911,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L829-L840" + "specLocation": "inference/_types/CommonTypes.ts#L839-L850" }, { "kind": "interface", @@ -168902,7 +168929,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L842-L980" + "specLocation": "inference/_types/CommonTypes.ts#L852-L990" }, { "kind": "interface", @@ -168980,7 +169007,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L759-L827" + "specLocation": "inference/_types/CommonTypes.ts#L769-L837" }, { "kind": "enum", @@ -168993,7 +169020,7 @@ "name": "CustomServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L989-L991" + "specLocation": "inference/_types/CommonTypes.ts#L999-L1001" }, { "kind": "interface", @@ -169011,7 +169038,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L993-L1007" + "specLocation": "inference/_types/CommonTypes.ts#L1003-L1017" }, { "kind": "enum", @@ -169033,7 +169060,7 @@ "name": "CustomTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L982-L987" + "specLocation": "inference/_types/CommonTypes.ts#L992-L997" }, { "kind": "interface", @@ -169081,7 +169108,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1021-L1043" + "specLocation": "inference/_types/CommonTypes.ts#L1031-L1053" }, { "kind": "enum", @@ -169094,7 +169121,7 @@ "name": "DeepSeekServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1045-L1047" + "specLocation": "inference/_types/CommonTypes.ts#L1055-L1057" }, { "kind": "interface", @@ -169235,7 +169262,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1070-L1104" + "specLocation": "inference/_types/CommonTypes.ts#L1080-L1114" }, { "kind": "enum", @@ -169248,7 +169275,7 @@ "name": "ElasticsearchServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1120-L1122" + "specLocation": "inference/_types/CommonTypes.ts#L1130-L1132" }, { "kind": "interface", @@ -169271,7 +169298,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1106-L1112" + "specLocation": "inference/_types/CommonTypes.ts#L1116-L1122" }, { "kind": "enum", @@ -169290,7 +169317,7 @@ "name": "ElasticsearchTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1114-L1118" + "specLocation": "inference/_types/CommonTypes.ts#L1124-L1128" }, { "kind": "interface", @@ -169336,7 +169363,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1124-L1150" + "specLocation": "inference/_types/CommonTypes.ts#L1134-L1160" }, { "kind": "enum", @@ -169349,7 +169376,7 @@ "name": "ElserServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1156-L1158" + "specLocation": "inference/_types/CommonTypes.ts#L1166-L1168" }, { "kind": "enum", @@ -169362,7 +169389,7 @@ "name": "ElserTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1152-L1154" + "specLocation": "inference/_types/CommonTypes.ts#L1162-L1164" }, { "kind": "enum", @@ -169375,7 +169402,7 @@ "name": "GoogleAiServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1183-L1185" + "specLocation": "inference/_types/CommonTypes.ts#L1193-L1195" }, { "kind": "interface", @@ -169423,7 +169450,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1160-L1176" + "specLocation": "inference/_types/CommonTypes.ts#L1170-L1186" }, { "kind": "enum", @@ -169439,7 +169466,7 @@ "name": "GoogleAiStudioTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1178-L1181" + "specLocation": "inference/_types/CommonTypes.ts#L1188-L1191" }, { "kind": "interface", @@ -169513,7 +169540,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1187-L1213" + "specLocation": "inference/_types/CommonTypes.ts#L1197-L1223" }, { "kind": "enum", @@ -169526,7 +169553,7 @@ "name": "GoogleVertexAIServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1233-L1235" + "specLocation": "inference/_types/CommonTypes.ts#L1243-L1245" }, { "kind": "interface", @@ -169560,7 +169587,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1215-L1224" + "specLocation": "inference/_types/CommonTypes.ts#L1225-L1234" }, { "kind": "enum", @@ -169582,7 +169609,7 @@ "name": "GoogleVertexAITaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1226-L1231" + "specLocation": "inference/_types/CommonTypes.ts#L1236-L1241" }, { "kind": "interface", @@ -169644,7 +169671,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1237-L1269" + "specLocation": "inference/_types/CommonTypes.ts#L1247-L1279" }, { "kind": "enum", @@ -169657,7 +169684,7 @@ "name": "HuggingFaceServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1290-L1292" + "specLocation": "inference/_types/CommonTypes.ts#L1300-L1302" }, { "kind": "interface", @@ -169691,7 +169718,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1271-L1281" + "specLocation": "inference/_types/CommonTypes.ts#L1281-L1291" }, { "kind": "enum", @@ -169713,7 +169740,7 @@ "name": "HuggingFaceTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1283-L1288" + "specLocation": "inference/_types/CommonTypes.ts#L1293-L1298" }, { "kind": "interface", @@ -170785,7 +170812,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1294-L1323" + "specLocation": "inference/_types/CommonTypes.ts#L1304-L1333" }, { "kind": "enum", @@ -170798,7 +170825,7 @@ "name": "JinaAIServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1353-L1355" + "specLocation": "inference/_types/CommonTypes.ts#L1363-L1365" }, { "kind": "enum", @@ -170817,7 +170844,7 @@ "name": "JinaAISimilarityType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1357-L1361" + "specLocation": "inference/_types/CommonTypes.ts#L1367-L1371" }, { "kind": "interface", @@ -170863,7 +170890,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1325-L1346" + "specLocation": "inference/_types/CommonTypes.ts#L1335-L1356" }, { "kind": "enum", @@ -170879,7 +170906,7 @@ "name": "JinaAITaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1348-L1351" + "specLocation": "inference/_types/CommonTypes.ts#L1358-L1361" }, { "kind": "enum", @@ -170901,7 +170928,7 @@ "name": "JinaAITextEmbeddingTask", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1363-L1368" + "specLocation": "inference/_types/CommonTypes.ts#L1373-L1378" }, { "kind": "interface", @@ -171059,7 +171086,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1370-L1397" + "specLocation": "inference/_types/CommonTypes.ts#L1380-L1407" }, { "kind": "enum", @@ -171072,7 +171099,7 @@ "name": "MistralServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1405-L1407" + "specLocation": "inference/_types/CommonTypes.ts#L1415-L1417" }, { "kind": "enum", @@ -171091,7 +171118,7 @@ "name": "MistralTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1399-L1403" + "specLocation": "inference/_types/CommonTypes.ts#L1409-L1413" }, { "kind": "interface", @@ -171178,7 +171205,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1409-L1451" + "specLocation": "inference/_types/CommonTypes.ts#L1419-L1461" }, { "kind": "enum", @@ -171191,7 +171218,7 @@ "name": "OpenAIServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1467-L1469" + "specLocation": "inference/_types/CommonTypes.ts#L1477-L1479" }, { "kind": "interface", @@ -171213,7 +171240,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1453-L1459" + "specLocation": "inference/_types/CommonTypes.ts#L1463-L1469" }, { "kind": "enum", @@ -171232,7 +171259,7 @@ "name": "OpenAITaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1461-L1465" + "specLocation": "inference/_types/CommonTypes.ts#L1471-L1475" }, { "kind": "interface", @@ -171620,13 +171647,16 @@ }, { "name": "completion" + }, + { + "name": "rerank" } ], "name": { "name": "TaskTypeAzureAIStudio", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L52-L55" + "specLocation": "inference/_types/TaskType.ts#L52-L56" }, { "kind": "enum", @@ -171642,7 +171672,7 @@ "name": "TaskTypeAzureOpenAI", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L57-L60" + "specLocation": "inference/_types/TaskType.ts#L58-L61" }, { "kind": "enum", @@ -171661,7 +171691,7 @@ "name": "TaskTypeCohere", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L62-L66" + "specLocation": "inference/_types/TaskType.ts#L63-L67" }, { "kind": "enum", @@ -171683,7 +171713,7 @@ "name": "TaskTypeCustom", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L68-L73" + "specLocation": "inference/_types/TaskType.ts#L69-L74" }, { "kind": "enum", @@ -171699,7 +171729,7 @@ "name": "TaskTypeDeepSeek", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L75-L78" + "specLocation": "inference/_types/TaskType.ts#L76-L79" }, { "kind": "enum", @@ -171712,7 +171742,7 @@ "name": "TaskTypeELSER", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L86-L88" + "specLocation": "inference/_types/TaskType.ts#L87-L89" }, { "kind": "enum", @@ -171731,7 +171761,7 @@ "name": "TaskTypeElasticsearch", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L80-L84" + "specLocation": "inference/_types/TaskType.ts#L81-L85" }, { "kind": "enum", @@ -171747,7 +171777,7 @@ "name": "TaskTypeGoogleAIStudio", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L90-L93" + "specLocation": "inference/_types/TaskType.ts#L91-L94" }, { "kind": "enum", @@ -171763,7 +171793,7 @@ "name": "TaskTypeGoogleVertexAI", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L95-L98" + "specLocation": "inference/_types/TaskType.ts#L96-L99" }, { "kind": "enum", @@ -171785,7 +171815,7 @@ "name": "TaskTypeHuggingFace", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L100-L105" + "specLocation": "inference/_types/TaskType.ts#L101-L106" }, { "kind": "enum", @@ -171820,7 +171850,7 @@ "name": "TaskTypeMistral", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L107-L111" + "specLocation": "inference/_types/TaskType.ts#L108-L112" }, { "kind": "enum", @@ -171839,7 +171869,7 @@ "name": "TaskTypeOpenAI", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L113-L117" + "specLocation": "inference/_types/TaskType.ts#L114-L118" }, { "kind": "enum", @@ -171855,7 +171885,7 @@ "name": "TaskTypeVoyageAI", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L119-L122" + "specLocation": "inference/_types/TaskType.ts#L120-L123" }, { "kind": "enum", @@ -171874,7 +171904,7 @@ "name": "TaskTypeWatsonx", "namespace": "inference._types" }, - "specLocation": "inference/_types/TaskType.ts#L124-L128" + "specLocation": "inference/_types/TaskType.ts#L125-L129" }, { "kind": "interface", @@ -172120,7 +172150,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1471-L1502" + "specLocation": "inference/_types/CommonTypes.ts#L1481-L1512" }, { "kind": "enum", @@ -172133,7 +172163,7 @@ "name": "VoyageAIServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1535-L1537" + "specLocation": "inference/_types/CommonTypes.ts#L1545-L1547" }, { "kind": "interface", @@ -172193,7 +172223,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1504-L1528" + "specLocation": "inference/_types/CommonTypes.ts#L1514-L1538" }, { "kind": "enum", @@ -172209,7 +172239,7 @@ "name": "VoyageAITaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1530-L1533" + "specLocation": "inference/_types/CommonTypes.ts#L1540-L1543" }, { "kind": "interface", @@ -172297,7 +172327,7 @@ } } ], - "specLocation": "inference/_types/CommonTypes.ts#L1539-L1577" + "specLocation": "inference/_types/CommonTypes.ts#L1549-L1587" }, { "kind": "enum", @@ -172310,7 +172340,7 @@ "name": "WatsonxServiceType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1585-L1587" + "specLocation": "inference/_types/CommonTypes.ts#L1595-L1597" }, { "kind": "enum", @@ -172329,7 +172359,7 @@ "name": "WatsonxTaskType", "namespace": "inference._types" }, - "specLocation": "inference/_types/CommonTypes.ts#L1579-L1583" + "specLocation": "inference/_types/CommonTypes.ts#L1589-L1593" }, { "kind": "request", @@ -173056,7 +173086,7 @@ } } }, - "description": "Create an inference endpoint.\n\nIMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)", + "description": "Create an inference endpoint.\n\nIMPORTANT: The inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Mistral, Azure OpenAI, Google AI Studio, Google Vertex AI, Anthropic, Watsonx.ai, or Hugging Face.\nFor built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models.\nHowever, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.\n\nThe following integrations are available through the inference API. You can find the available task types next to the integration name:\n* AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`)\n* Amazon Bedrock (`completion`, `text_embedding`)\n* Anthropic (`completion`)\n* Azure AI Studio (`completion`, 'rerank', `text_embedding`)\n* Azure OpenAI (`completion`, `text_embedding`)\n* Cohere (`completion`, `rerank`, `text_embedding`)\n* DeepSeek (`completion`, `chat_completion`)\n* Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)\n* ELSER (`sparse_embedding`)\n* Google AI Studio (`completion`, `text_embedding`)\n* Google Vertex AI (`rerank`, `text_embedding`)\n* Hugging Face (`chat_completion`, `completion`, `rerank`, `text_embedding`)\n* Mistral (`chat_completion`, `completion`, `text_embedding`)\n* OpenAI (`chat_completion`, `completion`, `text_embedding`)\n* VoyageAI (`text_embedding`, `rerank`)\n* Watsonx inference integration (`text_embedding`)\n* JinaAI (`text_embedding`, `rerank`)", "examples": { "InferencePutExample1": { "alternatives": [ @@ -173915,6 +173945,12 @@ "method_request": "PUT _inference/completion/azure_ai_studio_completion", "summary": "A completion task", "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"databricks\",\n \"endpoint_type\": \"realtime\"\n }\n}" + }, + "PutAzureAiStudioRequestExample3": { + "description": "Run `PUT _inference/rerank/azure_ai_studio_rerank` to create an inference endpoint that performs a rerank task.", + "method_request": "PUT _inference/rerank/azure_ai_studio_rerank", + "summary": "A rerank task", + "value": "{\n \"service\": \"azureaistudio\",\n \"service_settings\": {\n \"api_key\": \"Azure-AI-Studio-API-key\",\n \"target\": \"Target-URI\",\n \"provider\": \"cohere\",\n \"endpoint_type\": \"token\"\n }\n}" } }, "inherits": { diff --git a/output/typescript/types.ts b/output/typescript/types.ts index cab7b8f62c..51f7959779 100644 --- a/output/typescript/types.ts +++ b/output/typescript/types.ts @@ -13801,9 +13801,11 @@ export interface InferenceAzureAiStudioTaskSettings { temperature?: float top_p?: float user?: string + return_documents?: boolean + top_n?: integer } -export type InferenceAzureAiStudioTaskType = 'completion' | 'text_embedding' +export type InferenceAzureAiStudioTaskType = 'completion' | 'rerank' | 'text_embedding' export interface InferenceAzureOpenAIServiceSettings { api_key?: string @@ -14221,7 +14223,7 @@ export type InferenceTaskTypeAmazonBedrock = 'text_embedding' | 'completion' export type InferenceTaskTypeAnthropic = 'completion' -export type InferenceTaskTypeAzureAIStudio = 'text_embedding' | 'completion' +export type InferenceTaskTypeAzureAIStudio = 'text_embedding' | 'completion' | 'rerank' export type InferenceTaskTypeAzureOpenAI = 'text_embedding' | 'completion' diff --git a/specification/inference/_types/CommonTypes.ts b/specification/inference/_types/CommonTypes.ts index 4941cb9210..1e06afce52 100644 --- a/specification/inference/_types/CommonTypes.ts +++ b/specification/inference/_types/CommonTypes.ts @@ -567,10 +567,20 @@ export class AzureAiStudioTaskSettings { * This information can be used for abuse detection. */ user?: string + /** + * For a `rerank` task, return doc text within the results. + */ + return_documents?: boolean + /** + * For a `rerank` task, the number of most relevant documents to return. + * It defaults to the number of the documents. + */ + top_n?: integer } export enum AzureAiStudioTaskType { completion, + rerank, text_embedding } diff --git a/specification/inference/_types/TaskType.ts b/specification/inference/_types/TaskType.ts index 6daed0d281..a8a04bfc4e 100644 --- a/specification/inference/_types/TaskType.ts +++ b/specification/inference/_types/TaskType.ts @@ -51,7 +51,8 @@ export enum TaskTypeAnthropic { export enum TaskTypeAzureAIStudio { text_embedding, - completion + completion, + rerank } export enum TaskTypeAzureOpenAI { diff --git a/specification/inference/put/PutRequest.ts b/specification/inference/put/PutRequest.ts index 4554574e32..177589ee56 100644 --- a/specification/inference/put/PutRequest.ts +++ b/specification/inference/put/PutRequest.ts @@ -34,7 +34,7 @@ import { TaskType } from '@inference/_types/TaskType' * * AlibabaCloud AI Search (`completion`, `rerank`, `sparse_embedding`, `text_embedding`) * * Amazon Bedrock (`completion`, `text_embedding`) * * Anthropic (`completion`) - * * Azure AI Studio (`completion`, `text_embedding`) + * * Azure AI Studio (`completion`, 'rerank', `text_embedding`) * * Azure OpenAI (`completion`, `text_embedding`) * * Cohere (`completion`, `rerank`, `text_embedding`) * * DeepSeek (`completion`, `chat_completion`) diff --git a/specification/inference/put_azureaistudio/examples/request/PutAzureAiStudioRequestExample3.yaml b/specification/inference/put_azureaistudio/examples/request/PutAzureAiStudioRequestExample3.yaml new file mode 100644 index 0000000000..27b1f30212 --- /dev/null +++ b/specification/inference/put_azureaistudio/examples/request/PutAzureAiStudioRequestExample3.yaml @@ -0,0 +1,14 @@ +summary: A rerank task +description: Run `PUT _inference/rerank/azure_ai_studio_rerank` to create an inference endpoint that performs a rerank task. +method_request: 'PUT _inference/rerank/azure_ai_studio_rerank' +# type: "request" +value: |- + { + "service": "azureaistudio", + "service_settings": { + "api_key": "Azure-AI-Studio-API-key", + "target": "Target-URI", + "provider": "cohere", + "endpoint_type": "token" + } + }