Amazon Bedrock website screenshot

Amazon Bedrock

Amazon Bedrock is a fully managed AWS service that makes high-performing foundation models from leading AI companies available through a unified API for building generative AI applications. It supports text and image generation, conversational AI, model customization and fine-tuning, retrieval-augmented generation (RAG) via knowledge bases, autonomous agents, guardrails for responsible AI, and provisioned throughput for production workloads. The Bedrock APIs are AWS regional service endpoints accessed over HTTPS using AWS Signature Version 4 (SigV4) authentication, typically via the AWS SDKs.

4 APIs 0 Features
AIFoundation ModelsGenerative AILLMMachine LearningRAGAgentsResponsible AI

APIs

Amazon Bedrock API

The Amazon Bedrock management API provides operations for managing foundation models, custom models, model customization jobs, provisioned throughput, guardrails, knowledge base...

Amazon Bedrock Runtime API

The Amazon Bedrock Runtime API provides operations for invoking foundation models for inference, including synchronous, streaming, and conversational invocation patterns (Conver...

Amazon Bedrock Agent API

The Amazon Bedrock Agent API provides operations for managing and configuring autonomous AI agents, knowledge bases for RAG, data sources, and ingestion jobs. Authentication use...

Amazon Bedrock Agent Runtime API

The Amazon Bedrock Agent Runtime API provides operations for invoking Bedrock agents and retrieving content from knowledge bases for RAG applications. Authentication uses AWS Si...

Collections

GraphQL

Amazon Bedrock GraphQL API

Amazon Bedrock is a fully managed AWS service for accessing foundation models from AI21 Labs, Anthropic, Cohere, Meta, Mistral AI, and Stability AI. The API covers model invocat...

GRAPHQL

Pricing Plans

Rate Limits

Amazon Bedrock Rate Limits

2 limits

RATE LIMITS

FinOps

Semantic Vocabularies

Amazon Bedrock Context

20 classes · 8 properties

JSON-LD

JSON Structure

Bedrock Resource Structure

0 properties

JSON STRUCTURE

Example Payloads

Converse Example

2 fields

EXAMPLE

Invoke Model Example

2 fields

EXAMPLE

Resources

🔗
Website
Website
🔗
Documentation
Documentation
💰
Pricing
Pricing
📝
Signup
Signup
🌐
Portal
Portal
🌐
Console
Console
🔗
Login
Login
💬
Support
Support
🟢
StatusPage
StatusPage
📰
Blog
Blog
💬
FAQ
FAQ
🚀
GettingStarted
GettingStarted
📜
TermsOfService
TermsOfService
📜
PrivacyPolicy
PrivacyPolicy
👥
GitHubOrganization
GitHubOrganization

Sources

Raw ↑
opencollection: 1.0.0
info:
  name: Amazon Bedrock API
  version: '2023-04-20'
items:
- info:
    name: Foundation Models
    type: folder
  items:
  - info:
      name: Amazon Bedrock List foundation models
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/foundation-models
      params:
      - name: byProvider
        value: ''
        type: query
        description: Filter by model provider (e.g., amazon, anthropic, meta).
      - name: byOutputModality
        value: ''
        type: query
        description: Filter by output modality (TEXT, IMAGE, EMBEDDING).
      - name: byInferenceType
        value: ''
        type: query
        description: Filter by inference type (ON_DEMAND, PROVISIONED).
    docs: Lists the Amazon Bedrock foundation models that you can use. You can filter the results by provider, output modality,
      and inference type.
  - info:
      name: Amazon Bedrock Get foundation model details
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/foundation-models/:modelIdentifier
      params:
      - name: modelIdentifier
        value: ''
        type: path
        description: The model identifier (e.g., anthropic.claude-3-sonnet-20240229-v1:0).
    docs: Gets details about a specific Amazon Bedrock foundation model, including its capabilities, supported input and output
      modalities, and inference types.
- info:
    name: Model Customization
    type: folder
  items:
  - info:
      name: Amazon Bedrock List model customization jobs
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/model-customization-jobs
      params:
      - name: maxResults
        value: ''
        type: query
        description: Maximum number of results to return.
      - name: nextToken
        value: ''
        type: query
        description: Token for pagination.
      - name: statusEquals
        value: ''
        type: query
        description: Filter by job status.
    docs: Returns a list of model customization jobs that you have submitted.
  - info:
      name: Amazon Bedrock Create a model customization job
      type: http
    http:
      method: POST
      url: https://bedrock.{region}.amazonaws.com/model-customization-jobs
      body:
        type: json
        data: '{}'
    docs: Creates a fine-tuning job to customize a foundation model. You provide training data, a base model, and output configuration
      to produce a custom model tailored to your use case.
  - info:
      name: Amazon Bedrock Get model customization job details
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/model-customization-jobs/:jobIdentifier
      params:
      - name: jobIdentifier
        value: ''
        type: path
        description: The identifier of the customization job.
    docs: Retrieves the properties associated with a model customization job, including the status, creation time, and training
      metrics.
- info:
    name: Custom Models
    type: folder
  items:
  - info:
      name: Amazon Bedrock List custom models
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/custom-models
      params:
      - name: maxResults
        value: ''
        type: query
        description: Maximum number of results to return.
      - name: nextToken
        value: ''
        type: query
        description: Token for pagination.
      - name: baseModelArnEquals
        value: ''
        type: query
        description: Filter by base model ARN.
    docs: Returns a list of the custom models that you have created with the CreateModelCustomizationJob operation.
- info:
    name: Provisioned Throughput
    type: folder
  items:
  - info:
      name: Amazon Bedrock List provisioned model throughputs
      type: http
    http:
      method: GET
      url: https://bedrock.{region}.amazonaws.com/provisioned-model-throughput
      params:
      - name: maxResults
        value: ''
        type: query
        description: Maximum number of results to return.
      - name: nextToken
        value: ''
        type: query
        description: Token for pagination.
      - name: statusEquals
        value: ''
        type: query
        description: Filter by status.
    docs: Lists the provisioned throughputs that you have created.
  - info:
      name: Amazon Bedrock Create provisioned model throughput
      type: http
    http:
      method: POST
      url: https://bedrock.{region}.amazonaws.com/provisioned-model-throughput
      body:
        type: json
        data: '{}'
    docs: Creates a provisioned throughput with dedicated capacity for a foundation model or custom model. Provisioned throughput
      provides guaranteed inference performance.
bundled: true