Amazon Nova

Amazon Nova is a new generation of state-of-the-art foundation models from Amazon that deliver a compelling combination of accuracy, speed, and cost efficiency. Amazon Nova models are accessible through Amazon Bedrock and support text, image, video, speech understanding and generation across a range of model types: Nova Premier (1M context), Nova Pro, Nova Lite, Nova Micro (text-only), Nova Canvas (image generation), Nova Reel (video generation), and Nova Sonic (speech).

1 APIs 8 Features

Foundation ModelsGenerative AIImage GenerationMachine LearningMultimodalSpeechVideo Generation

APIs

Amazon Nova API

The Amazon Nova API provides programmatic access to Amazon Nova foundation models through Amazon Bedrock for text, image, and video generation, understanding, and reasoning task...

Collections

Amazon Nova on Amazon Bedrock Runtime API

OPEN

Pricing Plans

Amazon Nova Plans Pricing

3 plans

PLANS

Rate Limits

Amazon Nova Rate Limits

5 limits

RATE LIMITS

FinOps

Amazon Nova Finops

FINOPS

Features

Multiple Model Types

Seven specialized models: Nova Premier (1M context), Nova Pro, Nova Lite, Nova Micro (text), Nova Canvas (image), Nova Reel (video), Nova Sonic (speech).

Multimodal Input

Supports text, images, video, documents (PDF, CSV, DOCX, XLS, HTML), and speech as input modalities.

Long Context Windows

Up to 1 million token context window in Nova Premier; 300k tokens in Nova Pro and Lite; 128k in Nova Micro.

Streaming Responses

All understanding models support streaming for real-time interactive applications.

Batch Inference

All understanding models support batch processing for high-volume offline workloads.

Fine-Tuning

Nova Pro, Lite, and Micro support fine-tuning for domain-specific customization.

Model Distillation

Nova Premier can serve as a teacher model for distillation into Pro, Lite, and Micro.

Bedrock Integration

Natively integrated with Amazon Bedrock Knowledge Bases, Agents, Guardrails, Evaluations, and Prompt Flows.

Use Cases

Interactive Chat Interfaces

Build conversational AI applications with long context awareness using Nova Pro, Lite, or Micro.

Retrieval-Augmented Generation

Enhance knowledge retrieval accuracy by combining Nova models with Bedrock Knowledge Bases.

Agentic Applications

Build autonomous AI agents that reason and act using Nova models with Bedrock Agents.

Video and Document Analysis

Analyze video content and complex documents (PDF, DOCX, XLS) with Nova Pro and Premier.

Image Generation and Editing

Generate and edit high-quality images programmatically with Amazon Nova Canvas.

Video Generation

Create short video clips from text or image prompts using Amazon Nova Reel.

Voice Assistants

Build voice-enabled customer service and assistant applications with Nova Sonic speech model.

UI Workflow Automation

Automate UI interactions and screen navigation workflows using Nova vision capabilities.

Integrations

Amazon Bedrock

Primary access method; all Nova models are served through the Bedrock InvokeModel and InvokeModelWithResponseStream APIs.

Bedrock Knowledge Bases

Connect Nova models to structured and unstructured data sources for RAG applications.

Bedrock Agents

Orchestrate multi-step agentic workflows with tool use and memory using Nova as the reasoning engine.

Bedrock Guardrails

Apply safety guardrails to Nova Premier, Pro, and Lite model outputs for content filtering.

Bedrock Prompt Flows

Build visual prompt chaining workflows connecting Nova models with other services.

Bedrock Evaluations

Evaluate Nova model performance on custom benchmarks and safety criteria.

Amazon S3

Store and access training data, batch inference inputs/outputs, and generated media artifacts.

AWS IAM

Control access to Nova model invocations through fine-grained IAM policies and Bedrock model access settings.

Resources

GitHubOrganization

Sources

opencollection: 1.0.0
info:
  name: Amazon Nova on Amazon Bedrock Runtime API
  version: '2023-09-30'
request:
  auth:
    type: apikey
    key: Authorization
    value: '{{Authorization}}'
    placement: header
items:
- info:
    name: Inference
    type: folder
  items:
  - info:
      name: Invoke a Nova model
      type: http
    http:
      method: POST
      url: https://bedrock-runtime.{region}.amazonaws.com/model/:modelId/invoke
      headers:
      - name: Content-Type
        value: ''
      - name: Accept
        value: ''
      params:
      - name: modelId
        value: ''
        type: path
        description: Nova model identifier or inference profile id.
      body:
        type: json
        data: '{}'
    docs: 'Synchronously invoke a Nova model identified by its model ID

      (for example `amazon.nova-pro-v1:0`, `amazon.nova-canvas-v1:0`,

      or an inference profile id like `us.amazon.nova-lite-v1:0`).

      '
  - info:
      name: Invoke a Nova model with streaming output
      type: http
    http:
      method: POST
      url: https://bedrock-runtime.{region}.amazonaws.com/model/:modelId/invoke-with-response-stream
      params:
      - name: modelId
        value: ''
        type: path
      body:
        type: json
        data: '{}'
    docs: Invoke a Nova model with streaming output
  - info:
      name: Converse with a Nova model
      type: http
    http:
      method: POST
      url: https://bedrock-runtime.{region}.amazonaws.com/model/:modelId/converse
      params:
      - name: modelId
        value: ''
        type: path
      body:
        type: json
        data: '{}'
    docs: Use the unified Bedrock Converse API for chat-style interactions.
  - info:
      name: Converse with streaming
      type: http
    http:
      method: POST
      url: https://bedrock-runtime.{region}.amazonaws.com/model/:modelId/converse-stream
      params:
      - name: modelId
        value: ''
        type: path
      body:
        type: json
        data: '{}'
    docs: Converse with streaming
- info:
    name: Async
    type: folder
  items:
  - info:
      name: Start an asynchronous Nova invocation
      type: http
    http:
      method: POST
      url: https://bedrock-runtime.{region}.amazonaws.com/async-invoke
      body:
        type: json
        data: '{}'
    docs: Used for Nova Reel and other long-running generations.
  - info:
      name: Get async invocation status
      type: http
    http:
      method: GET
      url: https://bedrock-runtime.{region}.amazonaws.com/async-invoke/:invocationArn
      params:
      - name: invocationArn
        value: ''
        type: path
    docs: Get async invocation status
bundled: true