Nova Pro
Nova Pro is a first-generation Nova model built for high-accuracy multimodal tasks including financial document analysis, large-codebase review, and complex visual reasoning across a context of 300K tokens.
View API reference- Input and output price
- Input $0.80, Output $3.20, Per 1M tokens
- 24h uptime
- Loading AI Gateway uptime
import { streamText } from 'ai'
const result = streamText({ model: 'amazon/nova-pro', prompt: 'Why is the sky blue?'})Copy link to headingPlayground
Try out Nova Pro by Amazon. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Nova Pro
Nova Pro
Copy link to headingProviders
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
Copy link to headingUptime24 hours
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
Copy link to headingThroughput24 hours
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
Copy link to headingLatency24 hours
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Getting started
Call Nova Pro through AI Gateway with the AI SDK generateText and streamText functions, or through the OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages APIs by changing the base URL. AI Gateway authenticates the request and routes it to an available provider.
Install the AI SDK (pnpm add ai dotenv), create an API key from the API Keys page, and set it as AI_GATEWAY_API_KEY in your environment. Full setup is covered in the text generation quickstart.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', prompt: 'Why is the sky blue?', });
console.log(result.text);}
main().catch(console.error);Top-level parameters
The same Nova Pro request in each API format AI Gateway supports.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', system: 'You are a concise technical assistant.', prompt: 'Summarize the tradeoffs between static generation and SSR.', maxOutputTokens: 1024, temperature: 0.5, });
console.log(result.text);}
main().catch(console.error);Standard parameters like prompt, messages, temperature, and tools work as documented in the AI SDK docs. These are the parameters with model-specific behavior.
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Model ID in the form creator/model, e.g. amazon/nova-pro. AI Gateway routes the request to an available provider. |
maxOutputTokens | number | No | Hard cap on generated tokens. Nova Pro supports up to 8,192 output tokens. |
providerOptions | Record<string, JSONValue> | No | AI Gateway routing options under gateway, plus any provider-native options under the provider’s own namespace — see the table below. |
Input limits
| Input | Formats | Sources | Max count | Max size | Limits |
|---|---|---|---|---|---|
| Text | — | — | — | — | Prompt and response share the 300K-token context window |
| Image | — | URL, base64, Uint8Array | — | — | Sent as image parts in messages; counts as input tokens |
| — | URL, base64, Uint8Array | — | — | Sent as file parts in messages; counts as input tokens |
Provider options
Set AI Gateway routing options under providerOptions.gateway. For provider-specific options, pass them under the provider’s namespace as documented by the AI SDK.
Learn more in the AI SDK amazon-bedrock provider docs.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', prompt: 'Why is the sky blue?', providerOptions: { gateway: { only: ['bedrock'], }, }, });
console.log(result.text);}
main().catch(console.error);These AI Gateway routing options apply to every model. Provider-specific options pass through under the provider’s own namespace (for example providerOptions.anthropic) exactly as documented by the AI SDK.
| Parameter | Type | Required | Description |
|---|---|---|---|
providerOptions.gateway.only | string[] | No | Restrict routing to these provider slugs. Requests fail over only within the listed providers. |
providerOptions.gateway.order | string[] | No | Preferred provider order. Listed providers are tried first; unlisted providers remain available as fallbacks. |
providerOptions.gateway.sort | 'cost' | 'ttft' | 'tps' | No | Rank candidate providers by price, time to first token, or tokens per second instead of the default routing order. |
providerOptions.gateway.zeroDataRetention | boolean | No | Route only to providers with a zero-data-retention policy for this model. |
Routing across providers
AI Gateway serves the same model through multiple providers and fails over automatically. order expresses a preference while keeping every provider eligible; only is a hard allowlist — if none of the listed providers are available the request fails instead of falling back.
Options under a provider's own namespace (for example providerOptions.anthropic) are forwarded to that provider with the request. Providers ignore option namespaces that don't apply to them, so it is safe to set provider options alongside gateway routing options.
Image input
Send images alongside text as message parts. Images count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this image.' }, { type: 'image', image: 'https://example.com/photo.jpg' }, ], }, ], });
console.log(result.text);}
main().catch(console.error);PDF input
Attach PDFs as file parts. Their contents count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', messages: [ { role: 'user', content: [ { type: 'text', text: 'Summarize this document.' }, { type: 'file', mediaType: 'application/pdf', data: 'https://example.com/document.pdf', }, ], }, ], });
console.log(result.text);}
main().catch(console.error);Tool calling
Expose tools the model can call. Define each tool’s inputs with a Zod schema.
import { generateText, tool } from 'ai';import { z } from 'zod';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'amazon/nova-pro', prompt: 'What is the weather in San Francisco?', tools: { getWeather: tool({ description: 'Get the current weather for a location', inputSchema: z.object({ location: z.string() }), execute: async ({ location }) => ({ location, temperatureC: 18 }), }), }, });
console.log(result.text);}
main().catch(console.error);Copy link to headingMore models by Amazon
| Model |
|---|
| 1M | 0.4 s | 204 tps | $0.30/M | $2.50/M | Read$0.08/M | — | +1 |  | 12/02/2025 | |||||
| 300K | 0.3 s | 134 tps | $0.06/M | $0.24/M | — | | 12/03/2024 | |||||||
| 128K | 0.4 s | $0.04/M | $0.14/M | — | | 12/03/2024 | ||||||||
| — | $0.02/M | — | | 04/30/2024 |
Copy link to headingAbout Nova Pro
Nova Pro launched on December 3, 2024 as the accuracy-focused tier in Amazon's first Nova generation. Its purpose is clear: deliver the highest accuracy the first-generation architecture can achieve, particularly on structured and complex inputs where cheaper models produce unreliable results.
The accuracy advantage shows most clearly in three categories.
First, structured documents. Financial reports with dense tabular data, legal contracts with nested conditional clauses, and regulatory filings where a single misinterpreted figure cascades into downstream errors. Nova Pro parses tables correctly, maintains numerical consistency across a long context, and distinguishes between semantically similar but legally distinct phrasings.
Second, large codebases. Nova Pro reasons over up to 15,000 lines of code in a single request. Entire modules can be evaluated in context for automated code review, migration analysis, and refactoring assessments. The model tracks variable state, identifies subtle logic errors, and assesses architectural patterns across files.
Third, combined visual and textual evidence. An insurance platform can analyze a policy document alongside claim photos. A manufacturing quality system can compare product specs against inspection images. Nova Pro accepts text, images, and video in the same request. Its accuracy advantage over Nova Lite is most pronounced when visual content contains structured information (charts, schematics, annotated diagrams) that requires precise interpretation.
Copy link to headingWhat To Consider When Choosing a Provider
- Configuration: Nova Pro is the top input rate in the first-generation Nova lineup on AI Gateway. Watch the AI Gateway cost dashboard so Pro spend matches the error cost you avoid.
- Zero Data Retention: Zero Data Retention is available for this model. It is offered on a per-provider and model basis. See the documentation for details.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
Copy link to headingWhen to Use Nova Pro
Best for
- Financial document processing: Precision on tables, figures, and numerical data is non-negotiable
- Automated code review: Migration analysis for codebases up to 15,000 lines per request
- Multi-document reasoning: Compare contracts, cross-reference reports, and validate data against documentation
- Regulated industry analysis: Combined visual and textual analysis in insurance, healthcare, and legal workflows
Consider alternatives when
- Routine classification or extraction: Nova Lite handles multimodal inputs at a fraction of the cost when Pro-level accuracy isn't required
- Extended thinking or web grounding: Nova 2 Lite provides second-generation agentic and reasoning capabilities
- Text-only workloads: Nova Micro is significantly cheaper and faster for pure text processing
Copy link to headingConclusion
Nova Pro is for requests where a wrong answer costs more than the extra per-token spend. It targets structured documents, large codebases, and complex multimodal inputs within the first-generation Nova lineup. Route your hardest problems here and send the rest to cheaper models.
Copy link to headingFrequently Asked Questions
What makes Nova Pro more accurate than Nova Lite on financial documents?
Nova Pro is tuned for precision on structured inputs, including tables with dense numerical data, nested clauses, and cross-referenced figures. Nova Lite may approximate values or miss conditional relationships in complex documents. Pro maintains consistency across the full context.
How large of a codebase can Nova Pro analyze in one request?
Up to 15,000 lines of code per request within the context of 300K tokens. For larger codebases, split across multiple requests or focus on specific modules.
When does the cost premium over Nova Lite pay for itself?
Whenever the cost of an incorrect output exceeds the per-token price difference. In financial compliance, legal review, and code quality assessment, a single error caught by Pro that Lite would have missed can justify hundreds of requests at the higher rate.
Can Nova Pro handle complex reasoning tasks like multi-step proofs or planning?
Nova Pro prioritizes accuracy within its generation but doesn't include extended thinking or configurable reasoning budgets. For deliberate multi-step reasoning, use Nova 2 Lite, a second-generation model with those capabilities.
Is Nova Pro suitable as the default model in a high-volume pipeline?
Usually not as a blanket default. Its input rate is well above Nova Lite's. Most teams route easy traffic to Lite or Micro and send only high-risk work to Pro.
What regulated-industry workflows benefit most from Nova Pro?
Insurance claims processing that combines policy documents with photographic evidence, healthcare record analysis where numerical lab values must be parsed precisely, and legal contract comparison where subtle phrasing differences have material consequences.