FlagshipAmazon

Nova Premier

Name: Nova Premier
Availability: InStock
Author: Amazon

Nova Premier is Amazon's flagship multimodal model with text and image capabilities, featuring a 1M token context window for complex reasoning tasks.

Context 1.0M

Tier Flagship

Modalities text, image

Input from

$1.25 / 1M tokens

across 2 providers

Compare Prices

API Pricing

Cheapest on Amazon AWS — 40% below avg

Provider	Input / 1M	Output / 1M	Speed	TTFT	Updated
Amazon AWSBatch	$1.25	$6.25	42.0 t/s	1.3s	4/14/2026
Amazon AWS	$2.50	$12.50	42.0 t/s	1.3s	4/14/2026
OpenRouter	$2.50	$12.50	42.0 t/s	1.3s	4/14/2026

Prices updated daily. Last check: 4/14/2026

Model Details

General

Creator: Amazon
Family: Nova
Tier: Flagship
Context Window: 1.0M
Modalities: Text, Image

Capabilities

Tool Calling: No
Open Source: No

Strengths & Limitations

1M token context window supports very long document processing
Multimodal capabilities handle both text and image inputs
49.12 tokens per second output speed for production workloads
Amazon's flagship model with advanced reasoning capabilities
Integration with AWS ecosystem and services
Suitable for complex enterprise multimodal applications
Large context enables batch processing of multiple documents

No tool calling or function execution support
Proprietary model with no open source availability
916ms time to first token is slower than some competitors
Limited to text and image modalities only
No audio or video processing capabilities

Key Features

•1M token context window

•Text and image input processing

•Multimodal reasoning capabilities

•49.12 tokens per second generation speed

•AWS cloud integration

•Streaming response support

•Batch document processing

•Enterprise-grade availability

About Nova Premier

Nova Premier is Amazon's flagship model in the Nova family, designed for high-complexity multimodal tasks requiring extensive reasoning capabilities. As Amazon's top-tier offering, it represents the company's most advanced language model technology for enterprise and developer applications. The model supports both text and image inputs with a substantial 1M token context window, enabling processing of very long documents, extensive conversations, or large batches of content in a single request. Performance benchmarks show Nova Premier generates 49.12 output tokens per second with a time to first token of 916 milliseconds, indicating solid throughput for production workloads. Nova Premier competes directly with other flagship models like Claude Opus 4.6 and GPT-5.4, targeting use cases that demand sophisticated reasoning over multimodal content. Its positioning as Amazon's premier offering makes it suitable for complex enterprise applications where the combination of vision capabilities and extended context processing is essential.

Common Use Cases

Nova Premier is designed for complex enterprise applications requiring sophisticated multimodal reasoning over large amounts of content. The 1M token context window makes it particularly suitable for analyzing lengthy documents with embedded images, processing multiple files simultaneously, or maintaining extended conversational context. Common applications include document analysis workflows, visual content understanding at scale, comprehensive research synthesis, and enterprise knowledge management systems where both text and visual information must be processed together. Its flagship positioning makes it appropriate for mission-critical applications where advanced reasoning capabilities justify the computational cost.

Frequently Asked Questions

How much does Nova Premier cost per million tokens?

Nova Premier pricing varies by provider and usage type. Check the pricing table above for current rates across all available providers offering Nova Premier access.

What is Nova Premier best used for?

Nova Premier excels at complex multimodal tasks requiring extensive context processing, such as analyzing long documents with images, batch processing multiple files, visual content understanding, and enterprise applications needing sophisticated reasoning over both text and visual information.

Does Nova Premier support function calling or tool use?

No, Nova Premier does not currently support tool calling or function execution capabilities. It focuses on multimodal reasoning and content generation rather than external tool integration.