iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
For a Nova 2 project that has to run now, start with Nova 2 Lite. It is the best-documented model in the family: AWS publishes a Bedrock model card, a service card, and a developer guide for it, and it can be called through the Converse and InvokeModel operations using the model ID amazon.nova-2-lite-v1:0. Nova 2 Pro and Nova 2 Omni were announced as previews on December 2, 2025. The AWS material behind this guide does not establish whether either has left preview, what its API surface is, or where it can be called. Treat both as candidates to verify, not options to build on yet.
The details below come from AWS’s model card, service card, developer guide, and launch posts dated December 2025. AWS revises these pages without notice, so confirm figures on the live pages before you commit to a design.
The three models at a glance
The table separates what AWS has stated from what it has not. “Not stated” means the cited AWS page or announcement does not give that value. It does not mean the capability is absent.
| Item | Nova 2 Lite | Nova 2 Pro | Nova 2 Omni |
|---|---|---|---|
| Status in AWS material | Listed as active on Bedrock; model launch date December 2, 2025 (model card) | Preview, announced December 2, 2025 | Preview, announced December 2, 2025 |
| Positioning | Everyday multimodal work such as document processing, customer support, and automation | Highly complex, multistep tasks such as multi-document analysis, video reasoning, and software migrations | All-in-one multimodal reasoning and image generation |
| Inputs | Text, image, video (model card) | Not stated (December 2, 2025 announcement) | Text, image, video, speech (December 2, 2025 announcement) |
| Outputs | Text (model card) | Not stated (December 2, 2025 announcement) | Text and image (December 2, 2025 announcement) |
| Context window | 1M tokens (model card) | Not stated (December 2, 2025 announcement) | 1M tokens (December 2, 2025 announcement) |
| Maximum output | 64K tokens (model card) | Not stated (December 2, 2025 announcement) | Not stated (December 2, 2025 announcement) |
| Model ID | amazon.nova-2-lite-v1:0 |
Not stated (December 2, 2025 announcement) | Not stated (December 2, 2025 announcement) |
| Bedrock API support | Converse and InvokeModel (model card and developer guide) | Not stated (December 2, 2025 announcement) | Not stated (December 2, 2025 announcement) |
| Early access path as announced | Not applicable to the active model listing | Nova Forge customers; others contact their AWS account team | Nova Forge customers could request early access |
Nova 2 Lite: what AWS documents
Intended workloads
AWS positions Lite for simple automation, document processing, and customer support. Its service card adds multi-turn interactions, retrieval-augmented generation, and agentic workflows. The December 2025 announcement also names customer service chatbots, business process automation, software engineering, and business intelligence and research. These are vendor-described uses. The AWS material contains no independent evaluation of how Lite performs in them, so test against your own workload before relying on any of them.
#1 Best Overall
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
Lifecycle
The Bedrock model card lists Lite as active, with an “EOL no sooner than” date of December 2, 2026. That is the earliest end-of-life date AWS allows, not a scheduled retirement. At the time of writing it is less than two months away, so any production deployment needs a migration path for the period after that date.
Calling Lite through Amazon Bedrock
The Amazon Nova 2 Developer Guide lists three prerequisites: an AWS account with Bedrock access, IAM permissions that allow model invocation, and an installed AWS SDK. With those in place:
Rank #2
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
- Choose the Region you will deploy to, then choose the matching model identifier from the table below.
- Choose the API. Converse uses a single message format across models. InvokeModel is the lower-level call. The model card lists both and recommends the
bedrock-runtimeendpoint for new applications where possible. - Send test requests from a sandbox account and check the output, token usage, and latency for your own prompts.
- Check the feature matrix for every feature your design depends on (see below).
Model identifiers and inference prefixes
| Identifier | Where AWS shows it | Use |
|---|---|---|
amazon.nova-2-lite-v1:0 |
Model ID on the Bedrock model card | Base model ID; confirm on the regional availability page which identifier your Region accepts |
us.amazon.nova-2-lite-v1:0 |
Developer guide example for a U.S. Region | U.S. cross-Region inference |
global.amazon.nova-2-lite-v1:0 |
Boto3 Converse example in AWS’s Lite announcement (sample Region us-east-1) | Global cross-Region inference |
| Other geography-prefixed ID | Developer guide, for access outside the U.S.; exact prefixes not stated in the guide excerpt | Confirm the exact prefix on the regional availability page |
Minimal Converse request
The sketch below uses the standard Boto3 Converse request shape with the U.S. inference ID. It has not been run against a live account for this article, so test it in a sandbox first and replace the prompt and parameters with your own.
Free tools Windows power users keep installed
One-click scans. No signup required.
import boto3
client = boto3.client("bedrock-runtime", region_name="us-east-1")
response = client.converse(
modelId="us.amazon.nova-2-lite-v1:0",
messages=[
{
"role": "user",
"content": [{"text": "Summarize the termination clause in the text below."}],
}
],
inferenceConfig={"maxTokens": 512, "temperature": 0.3},
)
print(response["output"]["message"]["content"][0]["text"])Feature support on Lite
- Supported through
bedrock-runtime, according to the model card: response streaming, Guardrails, prompt optimization, and client-side tool calling. - Marked unsupported on the model card: structured outputs, Agents, and other listed Bedrock features.
Confirm the matrix for your target endpoint before assuming a feature works with Lite.
Rank #3
- Meet Echo Dot Max: Experience rich room-filling sound that automatically adapts to your space and fine-tunes playback. Features a built-in smart home hub and Omnisense technology for highly personalized experiences.
- Music to your ears: With nearly 3x the bass versus Echo Dot (2022 release), it fits beautifully in any space, delivering your personal sound stage with deep bass and enhanced clarity. Listen to streaming services, such as Amazon Music, Apple Music, Spotify, and SiriusXM. Encore!
- Do more with device pairing: Connect compatible Echo smart speakers and smart displays in different rooms, or pair with a second Echo Dot Max to enjoy even richer sound
- Simple smart home control: Set routines, pair and control lights, locks, and thousands of smart home devices that work with Alexa without needing a separate smart home hub. With Omnisense technology, you can activate routines via temperature or presence detection.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot Max doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
Reasoning: extended thinking on Lite
Extended thinking is off by default. You can turn it on and set reasoning effort to low, medium, or high. The exact request fields are in the Nova 2 developer guide.
- Off (default): the model returns a standard response.
- Low, medium, or high: the model reasons before answering. AWS says the reasoning content is redacted from the output, and the reasoning tokens are billable.
The cost and observability trade-off
Reasoning tokens are billed even though you cannot read them. That makes per-request cost harder to predict, makes it harder to explain why an answer came out as it did, and leaves reviewers with a result they cannot inspect. Before you move to a higher effort level, compare outputs with and without it on a fixed set of your own tasks, and record token counts for each setting.
Rank #4
- Your favorite music and content – Play music, audiobooks, and podcasts from Amazon Music, Apple Music, Spotify and others or via Bluetooth throughout your home.
- Alexa is happy to help – Ask Alexa for weather updates and to set hands-free timers, get answers to your questions and even hear jokes. Need a few extra minutes in the morning? Just tap your Echo Dot to snooze your alarm.
- Keep your home comfortable – Control compatible smart home devices with your voice and routines triggered by built-in motion or indoor temperature sensors. Create routines to automatically turn on lights when you walk into a room, or start a fan if the inside temperature goes above your comfort zone.
- Do more with device pairing – Fill your home with music using compatible Echo devices in different rooms, or create a home theatre system with Fire TV.
- Say goodbye to drop-offs and buffering - With eero Built-in, Echo Dot doubles as a mesh wifi extender, adding up to 1,000 sq. ft. of wifi coverage to your existing eero network.
Nova 2 Pro and Nova 2 Omni: what was announced
Nova 2 Pro (preview, announced December 2, 2025)
AWS announced Pro as a preview for highly complex, multistep tasks, citing multi-document analysis, video reasoning, and software migrations as examples. At announcement, early access went to Nova Forge customers. Other customers were told to contact their AWS account team. The December 2025 material gives no Pro model ID, API surface, Region list, or quota, so this guide does not supply one.
Nova 2 Omni (preview, announced December 2, 2025)
AWS described Omni as an all-in-one model for multimodal reasoning and image generation. The announcement listed text, image, video, and speech inputs; text and image outputs; image generation and editing; and speech transcription. It stated a 1M-token context window, support for more than 200 languages in text processing, and 10 languages for speech input. Nova Forge customers could request early access at launch. The same material does not establish current interfaces or regional access.
Best Value
- MEET ECHO SPOT - A sleek smart alarm clock with Alexa and big vibrant sound. Ready to help you wake up, wind down, and so much more.
- CUSTOMIZABLE SMART CLOCK - See time, weather, and song titles at a glance, control smart home devices, and more. Personalize your display with your favorite clock face and fun colors.
- BIG VIBRANT SOUND - Enjoy rich sound with clear vocals and deep bass. Just ask Alexa to play music, podcasts, and audiobooks. See song titles and touch to control your music.
- EASE INTO THE DAY - Set up an Alexa routine that gently wakes you with music and gradual light. Glance at the time, check reminders, or ask Alexa for weather updates.
- KEEP YOUR HOME COMFORTABLE - Control compatible smart home devices. Just ask Alexa to turn on lights or touch the screen to dim. Create routines that use motion detection to turn down the thermostat as you head out or open the blinds when you walk into a room.
Verifying Pro and Omni before you plan around them
- Check the Amazon Bedrock console and the current Amazon Nova documentation for each model’s status (preview or generally available), model ID, and supported Regions.
- Confirm which APIs the model supports and which input and output modalities are enabled on your endpoint.
- Check Service Quotas for the default limits that apply to your account and Region.
- If you need preview access, contact your AWS account team, or use the Nova Forge route both announcements described.
Cost, quotas, and vendor performance claims
On-demand inference is billed per input and output token, and default quotas vary by model and Region. This guide does not list a per-token rate, because rates change. Use the live Amazon Bedrock pricing page for current figures, and Service Quotas for current limits and increase requests.
The AWS material includes two performance-related statements from outside AWS’s own benchmarks. Both are reported as stated by their source:
- AWS vendor claim: the Nova product page describes AWS Transform evaluations of Nova 2 Lite that found up to a 60% improvement in tool-calling efficiency for code modernization. This is AWS’s claim about one workload, not a general benchmark result.
- Customer testimonial: Trey Pezzetti, AI Product Manager at PGA TOUR, said that with Nova 2 Lite on Amazon Bedrock, “we can cover every shot, every player, every hole, every week at a cost that makes it viable to run continuously and build on top of.” This is one customer’s account, not a comparative cost study.
Customization and built-in tools
The Lite service card describes two built-in tools: Amazon Nova Grounding for web access, and a Code Interpreter for Python calculations. The same card describes supervised fine-tuning and reinforcement fine-tuning on Amazon Bedrock and SageMaker AI. The Nova product page adds broader customization pathways and deployment of custom models on SageMaker AI. Whether a given tool or customization path applies to Lite in your Region, and under what service restrictions, is set in the current service documentation. Confirm both before you design around them.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Evaluating a Nova 2 deployment
No independent benchmark results appear in the AWS material for any of the three models. Speed, quality, reliability, and value claims therefore remain AWS’s unless your own testing confirms them. AWS’s AI Service Card for Lite offers a useful framework for that testing. It asks you to define:
- the specific business problem and the stakeholders who will judge the output;
- the workflow, the system inputs and outputs, and the sources of variation in those inputs;
- the impact of errors, which determines how much verification each output needs.
The service card may not apply to customized SageMaker models or to models hosted on nova.amazon.com. If your deployment falls into either category, do not assume the card describes its risks.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
