A

Amazon Nova 2 Lite v1

Amazon
Vision
1M
Context
$0.3000
Input /1M
$2.50
Output /1M
N/A
Max Output
Run API Call Open in Chat

Amazon Nova 2 Lite v1

Overview

Nova 2 Lite is a fast, cost-effective reasoning model designed for everyday workloads that can process text, images, and videos to generate text. The model demonstrates standout capabilities in:

  • Document processing
  • Video information extraction
  • Code generation
  • Grounded answers
  • Multi-step agentic workflows

Model ID: amazon/nova-2-lite-v1
Created: December 2, 2025

Specifications

Context & Capacity

  • Context Length: 1,000,000 tokens
  • Maximum Completion Tokens: 65,535
  • Input Modalities: Text, image, video, file
  • Output Modalities: Text only

Reasoning

  • Reasoning Support: Yes
  • Supports reasoning and include_reasoning parameters

Pricing

Type Price
Context Window 1,000,000 tokens
Input Tokens $0.30 per million
Output Tokens $2.50 per million
Image Processing No additional charge

Performance

Recent Usage Metrics (December 23, 2025)

  • Requests: 4,427
  • Prompt Tokens: 17.56M
  • Completion Tokens: 3.6M
  • Tool Calls: 209

The LangMart page does not provide information about related or competing model variants at this time. For additional Nova model variants, refer to the Amazon Nova model family on LangMart.

Providers

Amazon Bedrock (sole provider)

  • Model Identifier: us.amazon.nova-2-lite-v1:0

Parameters

Supported Parameters

  • reasoning - Enable reasoning mode
  • include_reasoning - Include reasoning in output
  • max_tokens - Maximum output tokens
  • temperature - Sampling temperature control
  • top_p - Nucleus sampling parameter
  • top_k - Top-k sampling parameter
  • stop - Stop sequences
  • tool_choice - Tool selection (supports: literal none, auto, required, function types)
  • tools - Tool definitions (fully supported)

Input/Output Features

  • Multipart Input: Enabled
  • Tool Parameters: Fully supported
  • File URL Processing: Unavailable
  • Base64 Video Input: Supported