A multi-layer server block with red glowing strips, representing Amazon's own foundation model family.
Amazon Nova is a stacked family of models, each tier tuned for a different balance of speed, cost, and capability.

Amazon Nova is Amazon’s own family of foundation models , announced in December 2024. The family covers text, image, and video, and every model is delivered through Amazon Bedrock , the managed service that exposes many providers behind a single API. Nova solves a specific procurement problem for AWS customers: it gives them a first-party model line that AWS prices, supports, and integrates directly, rather than depending only on third-party models hosted on the platform.

The idea behind a model family is tiering. Instead of one model, Amazon ships several, each set at a different point on the speed, cost, and capability curve. You pick the smallest model that still handles your task well, then move up a tier only when accuracy demands it.

Where Nova sits in the stack

Nova is not a standalone product with its own console. It is a set of models you call through Bedrock, alongside models from other providers such as Anthropic’s Claude .

Your application
Web or mobile app Backend service Sends prompts, receives completions
Access layer
Amazon Bedrock API Knowledge Bases Agents Guardrails
Understanding models
Nova Micro Nova Lite Nova Pro Nova Premier Text and multimodal input, text output
Creative models
Nova Canvas Nova Reel Image and video generation

The understanding models

These take text, and in most cases images and video, and return text.

  • Nova Micro is the text-only tier. It targets the lowest latency and lowest cost per call in the family. Use it for high-volume, simple tasks like classification or short extraction.
  • Nova Lite is a low-cost multimodal model. It accepts image, video, and text input and returns text. Amazon positions it for fast, real-time interactions.
  • Nova Pro is the more capable multimodal model. AWS points it at complex tasks and agentic workflows, and it supports function calling and tool use.
  • Nova Premier is the most capable tier in the family and is positioned as a teacher model for distilling smaller custom variants.

The creative models

These generate media rather than text.

  • Nova Canvas generates images from text or image prompts and includes editing features such as inpainting, outpainting, and background removal.
  • Nova Reel generates short videos from text or image prompts, with camera controls such as zoom and movement.

How to access it and typical use

You do not install Nova. You call it through Amazon Bedrock, so the prerequisite is an AWS account with Bedrock access enabled for the Nova models you want in your Region.

A typical text request flows like this.

Step 1 Enable access Request model access to the Nova tiers in the Bedrock console for your Region.
→
Step 2 Pick a tier Start with Micro or Lite. Move to Pro or Premier only if accuracy needs it.
→
Step 3 Call the API Send the model ID and prompt through the Bedrock Converse or InvokeModel API.
→
Step 4 Add context Wire in Knowledge Bases for retrieval, Agents for tool use, Guardrails for policy.

Because Nova is a first-party AWS line, it plugs into the rest of Bedrock without extra glue. You can ground a Nova model on your own documents with Bedrock Knowledge Bases, give it tools through Bedrock Agents, and enforce content policy with Bedrock Guardrails. Amazon also documents fine-tuning and model distillation for the understanding tiers, where a larger model such as Premier teaches a smaller, cheaper custom variant.

Common uses:

  • Document and image understanding with Lite or Pro, where a request mixes text and visual input.
  • High-volume text classification with Micro, where cost per call matters more than depth.
  • Agentic workflows with Pro, using function calling and tool use to act on external systems.
  • Marketing and product media with Canvas and Reel for generated images and short video.

How it compares

Nova competes on two fronts: against other models hosted inside Bedrock, and against first-party model families from other clouds and labs.

Amazon NovaClaude (Anthropic)Azure OpenAIGoogle Gemini
MakerAmazonAnthropicOpenAIGoogle
Primary accessAmazon BedrockBedrock, direct APIAzureGoogle Cloud
First-party to a cloudYes, AWSNoYes, AzureYes, GCP
ModalitiesText, image, videoText, imageText, imageText, image, audio, video
Best forAWS-native, cost-tiered workReasoning, long contextAzure shopsGCP shops

Inside Bedrock, the practical choice is often Nova versus Claude. Nova is Amazon’s cost-and-speed play across a wide task range. Claude tends to be the reach-for model when a task needs deeper reasoning or careful instruction following. See the wider field in the LLM landscape for 2026 .

When not to use it

  • You are not on AWS. Nova is delivered through Bedrock. If your stack lives on Azure or Google Cloud, a first-party family there fits your billing and IAM better.
  • You want a portable, cloud-neutral contract. Building on a first-party model ties you more tightly to one cloud. Weigh that lock-in against the integration benefits.
  • The task needs top-end reasoning. For the hardest reasoning or agentic tasks inside Bedrock, evaluate Claude against Nova on your own data before committing.
  • You need a capability Nova does not cover. Match the specific modality and context needs of your task to a model’s documented specs, and confirm current specs in the AWS docs rather than assuming.

Further reading

Sources