Create an Amazon Bedrock inference endpoint
- PUT /_inference/{task_type}/{amazonbedrock_inference_id}
Create an inference endpoint to perform an inference task with the amazonbedrock service.
You need to provide the access and secret keys only once, during the inference model creation. The get inference API does not retrieve your access or secret keys. After creating the inference model, you cannot change the associated key pairs. If you want to use a different access and secret key pair, delete the inference model and recreate it with the same name and the updated keys.
Required authorization
- Cluster privileges manage_inference
Request body
-
chunking_settingsobject -
The chunking configuration object. Applies only to the
text_embeddingtask type. Not applicable to thechat_completionandcompletiontask types. - Reference
-
max_chunk_sizenumber · default: 250 -
The maximum size of a chunk in words. This value cannot be lower than
20(forsentencestrategy) or10(forwordstrategy). This value should not exceed the window size for the associated model. -
overlapnumber · default: 100 -
The number of overlapping words for chunks. It is applicable only to a
wordchunking strategy. This value cannot be higher than half themax_chunk_sizevalue. -
sentence_overlapnumber · default: 1 -
The number of overlapping sentences for chunks. It is applicable only for a
sentencechunking strategy. It can be either1or0. -
separator_groupstring -
Only applicable to the
recursivestrategy and required when using it.Sets a predefined list of separators in the saved chunking settings based on the selected text type. Values can be
markdownorplaintext.Using this parameter is an alternative to manually specifying a custom
separatorslist. -
separatorsstring -
Only applicable to the
recursivestrategy and required when using it.A list of strings used as possible split points when chunking text.
Each string can be a plain string or a regular expression (regex) pattern. The system tries each separator in order to split the text, starting from the first item in the list.
After splitting, it attempts to recombine smaller pieces into larger chunks that stay within the
max_chunk_sizelimit, to reduce the total number of chunks generated. -
strategystring · default: sentence -
The chunking strategy:
sentence,word,noneorrecursive.- If
strategyis set torecursive, you must also specify:
max_chunk_size- either
separatorsorseparator_group
Learn more about different chunking strategies in the linked documentation.
- If
- Reference
-
servicerequired -
The type of service supported for the specified task type. In this case,
amazonbedrock. -
Values:
amazonbedrock -
service_settingsobjectrequired -
Settings used to install the inference model. These settings are specific to the
amazonbedrockservice. -
access_keystringrequired -
A valid AWS access key that has permissions to use Amazon Bedrock and access to models for inference requests.
-
modelstringrequired -
The base model ID or an ARN to a custom model based on a foundational model. The base model IDs can be found in the Amazon Bedrock documentation. Note that the model ID must be available for the provider chosen and your IAM user must have access to the model.
- External docs
-
providerstring -
The model provider for your deployment. Note that some providers may support only certain task types. Supported providers include:
amazontitan- available fortext_embeddingandcompletiontask typesanthropic- available forchat_completionandcompletiontask typesai21labs- available forchat_completionandcompletiontask typescohere- available forchat_completion,completionandtext_embeddingtask typesmeta- available forchat_completionandcompletiontask typesmistral- available forchat_completionandcompletiontask types
-
regionstringrequired -
The region that your model or ARN is deployed in. The list of available regions per model can be found in the Amazon Bedrock documentation.
- External docs
-
rate_limitobject -
This setting helps to minimize the number of rate limit errors returned from Amazon Bedrock. By default, the
amazonbedrockservice sets the number of requests allowed per minute to 240. -
requests_per_minutenumber -
The number of requests allowed per minute. By default, the number of requests allowed per minute is set by each service as follows:
alibabacloud-ai-searchservice:1000amazonbedrockservice:240anthropicservice:50azureaistudioservice:240azureopenaiservice and task typetext_embedding:1440azureopenaiservice and task typescompletionorchat_completion:120cohereservice:10000contextualaiservice:1000elasticservice and task typechat_completion:240fireworksaiservice:6000googleaistudioservice:360googlevertexaiservice:30000hugging_faceservice:3000jinaaiservice:2000llamaservice:3000mistralservice:240openaiservice and task typetext_embedding:3000openaiservice and task typecompletion:500openshift_aiservice:3000voyageaiservice:2000watsonxaiservice:120
-
secret_keystringrequired -
A valid AWS secret key that is paired with the
access_key. For informationg about creating and managing access and secret keys, refer to the AWS documentation. - External docs
-
task_settingsobject -
Settings to configure the inference task. These settings are specific to the task type you specified.
-
max_new_tokensnumber · default: 64 -
For
chat_completionandcompletiontasks, it sets the maximum number for the output tokens to be generated. -
temperaturenumber -
For
chat_completionandcompletiontasks, it is a number between 0.0 and 1.0 that controls the apparent creativity of the results. At temperature 0.0 the model is most deterministic, at temperature 1.0 most random. It should not be used iftop_portop_kis specified. -
top_knumber -
For
chat_completionandcompletiontasks, it limits samples to the top-K most likely words, balancing coherence and variability. It is only available for anthropic, cohere, and mistral providers. It is an alternative totemperature; it should not be used iftemperatureis specified. -
top_pnumber -
For
chat_completionandcompletiontasks, it is a number in the range of 0.0 to 1.0, to eliminate low-probability tokens. Top-p uses nucleus sampling to select top tokens whose sum of likelihoods does not exceed a certain value, ensuring both variety and coherence. It is an alternative totemperature; it should not be used iftemperatureis specified.
Response
-
chunking_settingsobject -
The chunking configuration object. Applies only to the
embedding,sparse_embeddingandtext_embeddingtask types. Not applicable to thererank,completion, orchat_completiontask types. -
max_chunk_sizenumber · default: 250 -
The maximum size of a chunk in words. This value cannot be lower than
20(forsentencestrategy) or10(forwordstrategy). This value should not exceed the window size for the associated model. -
overlapnumber · default: 100 -
The number of overlapping words for chunks. It is applicable only to a
wordchunking strategy. This value cannot be higher than half themax_chunk_sizevalue. -
sentence_overlapnumber · default: 1 -
The number of overlapping sentences for chunks. It is applicable only for a
sentencechunking strategy. It can be either1or0. -
separator_groupstring -
Only applicable to the
recursivestrategy and required when using it.Sets a predefined list of separators in the saved chunking settings based on the selected text type. Values can be
markdownorplaintext.Using this parameter is an alternative to manually specifying a custom
separatorslist. -
separatorsstring -
Only applicable to the
recursivestrategy and required when using it.A list of strings used as possible split points when chunking text.
Each string can be a plain string or a regular expression (regex) pattern. The system tries each separator in order to split the text, starting from the first item in the list.
After splitting, it attempts to recombine smaller pieces into larger chunks that stay within the
max_chunk_sizelimit, to reduce the total number of chunks generated. -
strategystring · default: sentence -
The chunking strategy:
sentence,word,noneorrecursive.- If
strategyis set torecursive, you must also specify:
max_chunk_size- either
separatorsorseparator_group
Learn more about different chunking strategies in the linked documentation.
- If
- Reference