Amazon: Amazon Nova 2 Lite
nova-2-lite- Input / Output
- $0.30 / $2.50
- Context
- 1M
- Modalities
- Chat
- Capabilities
- Tool useReasoning
About Amazon Nova 2 Lite
Amazon Nova 2 Lite is Amazon's fast reasoning model for everyday workloads such as document processing, conversational applications, and tool use. It supports extended thinking at three effort levels, off by default, and works on text input. Amazon includes web grounding and a code interpreter as built-in tools. It is the efficient tier of the Nova 2 family.
Where it works well
- Extended thinking has low, medium, and high budgets, so reasoning effort can be set per request.
- Amazon reports improved document understanding for long text with complex structure, such as reports, contracts, and multi-page material.
- The long context fits large documents, codebases, or transcripts in one request.
- Tool use is a design focus; Amazon positions the model for multi-step agent workflows with several tools.
When to choose another model
- Text input only, so pictures and scanned pages must be converted to text first.
- As the lite tier, it is not Amazon's most capable Nova model for the hardest reasoning tasks.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
Text to textChat Completions APIPOST/v1/chat/completionscurl https://api.tokenlab.sh/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-xxx" \ -d '{ "model": "nova-2-lite", "messages": [ {"role": "user", "content": "Hello!"} ] }'
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Default spec
per 1M tokens- Official price
- Input $0.30 / Output $2.50 / Cache read $0.075
- Verified price
- —
- Discount
- —
Cache read
- Official price
- $0.075
- Verified price
- —
- Discount
- —
| Official priceper 1M tokens | Verified priceper 1M tokens | Discount | |
|---|---|---|---|
| Default spec | Input $0.30 / Output $2.50 / Cache read $0.075 | — | — |
| Prompt cache pricing | |||
| Cache read | $0.075 | — | — |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open Amazon Nova 2 Lite in Console with a prompt ready to edit or send.
Help me try nova-2-lite with a short message at /v1/chat/completions. Show the reply, latency, and cost.
Use cases
Best for- Reasoning
- Code
Document processing
Extract fields and summaries from long contracts, invoices, and reports with tables.
Customer support agents
Reason through a support case, call lookup tools, and write a reply.
Code assistance
Help with coding tasks, such as reviewing a function or planning a refactor, where a modest reasoning budget improves accuracy.
Cost-conscious reasoning
Switch thinking on for hard questions and leave it off for routine chat turns.
Prompt examples
Read this 80-page annual report and list every risk factor mentioned, grouped by theme, with page references.
A customer reports a double charge. Use the order lookup tool, then draft a reply with the next steps.
Think carefully: plan a migration of this three-table schema with no downtime and list the order of steps.
This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.
FAQ
What is Amazon Nova 2 Lite best for?
Everyday production workloads: document processing, conversational applications, and agents that call tools. Its extended thinking can be raised when a task needs more reasoning and left off when speed matters.
How does extended thinking work in Nova 2 Lite?
It is off by default. When enabled, you pick low, medium, or high effort, and the model reasons internally before replying. Amazon notes the reasoning content is redacted in the output.
Does Nova 2 Lite have a long context?
Yes. Amazon documents a one million token window for the model, which allows very long documents in a single request. The page above shows the current exact limit.
What built-in tools does it have?
Amazon lists web grounding for up-to-date information and a code interpreter that runs Python for calculations. Your own function tools are also supported through the chat interface.
Do prompts written for other chat models carry over?
Yes. Nova 2 Lite takes the standard chat message format, so prompts written for other chat models carry over with little change. Adjust the thinking effort to suit the task.
How much does Amazon Nova 2 Lite cost?
On TokenLab, Amazon Nova 2 Lite costs Input $0.30 / Output $2.50 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.
What are the context window and output limit of Amazon Nova 2 Lite?
Amazon Nova 2 Lite accepts up to 1,000,000 tokens of context.
Which endpoint should Amazon Nova 2 Lite use?
Use https://api.tokenlab.sh/v1/chat/completions for Amazon Nova 2 Lite. The request example below shows the matching code shape.
Which operations does Amazon Nova 2 Lite support?
Amazon Nova 2 Lite supports Text to text. Select an operation above to see its endpoint and request example.
Compare Amazon Nova 2 Lite
Sources
Reviewed Oct 2, 2026
