Amazon Nova Lite
Amazon Nova Lite is a multimodal text generation model from Amazon with a 300,000 token context window.
Amazon
Amazon BedrockFast multimodal text generation at low cost
Amazon Nova Lite is a text generation model developed by Amazon and released in December 2024, available through Amazon Bedrock. It supports multimodal inputs and offers a 300,000 token context window, making it suited for tasks that involve long documents or extended conversations. The model is tagged as a flagship offering within the Nova family and is designed to balance speed and cost efficiency.
Amazon Nova Lite is positioned for use cases where low latency and low cost are priorities, such as summarization, classification, and content generation at scale. Its large context window allows it to process substantial amounts of text in a single request, which is useful for document analysis and retrieval-augmented generation workflows. The model is accessible via Amazon Bedrock, meaning it integrates with AWS infrastructure and IAM-based access controls.
What Amazon Nova Lite supports
Large Context Window
Supports up to 300,000 tokens of context in a single request, enabling processing of long documents, transcripts, or multi-turn conversations without truncation.
Multimodal Input
Accepts multiple input modalities beyond plain text, allowing integration of images and other content types within a single prompt.
High-Speed Inference
Designed for low-latency responses, making it suitable for real-time applications and high-throughput pipelines.
Cost-Efficient Generation
Tagged as a low-cost model, making it practical for large-scale deployments where per-token cost is a key consideration.
Text Generation
Generates coherent text for tasks such as summarization, classification, question answering, and content drafting, with a maximum response size of 5,000 tokens.
AWS Bedrock Integration
Deployed via Amazon Bedrock, enabling access through standard AWS APIs with IAM-based authentication and no separate model hosting required.
Ready to build with Amazon Nova Lite?
Get Started FreeBenchmark scores
Scores represent accuracy — the percentage of questions answered correctly on each test.
| Benchmark | What it tests | Score |
|---|---|---|
| MMLU-Pro | Expert knowledge across 14 academic disciplines | 59.0% |
| GPQA Diamond | PhD-level science questions (biology, physics, chemistry) | 43.3% |
| MATH-500 | Undergraduate and competition-level math problems | 76.5% |
| AIME 2024 | American math olympiad problems | 10.7% |
| LiveCodeBench | Real-world coding tasks from recent competitions | 16.7% |
| HLE | Questions that challenge frontier models across many domains | 4.6% |
| SciCode | Scientific research coding and numerical methods | 13.9% |
Common questions about Amazon Nova Lite
What is the context window size for Amazon Nova Lite?
Amazon Nova Lite supports a context window of 300,000 tokens, allowing large volumes of text or multimodal content to be processed in a single request.
What is the maximum response size?
The model has a maximum response size of 5,000 tokens per generation.
How is Amazon Nova Lite accessed?
It is available through Amazon Bedrock under the model ID amazon.nova-lite-v1:0, using standard AWS authentication and API access.
What input types does Amazon Nova Lite support?
Amazon Nova Lite is tagged as multimodal, meaning it supports inputs beyond plain text, including images, in addition to standard text prompts.
When was Amazon Nova Lite released?
Amazon Nova Lite was released in December 2024 and became available on MindStudio on December 6, 2024.
Parameters & options
Explore similar models
AmazonAmazon Nova Micro
Amazon Nova Micro is a text-only generation model from Amazon with a 128,000-token context window and a 5,000-token response limit.
AmazonAmazon Nova 2 Lite
Amazon Nova 2 Lite is a text and image-input model from Amazon with a 1,000,000-token context window.
AmazonAmazon Nova Pro
Amazon Nova Pro is a text generation model from Amazon with a 300,000-token context window, available via Amazon Bedrock.
DeepSeekDeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 is a text generation model from DeepSeek with a 1,048,576-token context window and configurable reasoning effort.
QwenQwen3.8 2.4T
Qwen3.8 2.4T is a mixture-of-experts text generation model from Qwen with a 262,144 token context window.
GoogleGemini 3.5 Flash Lite
Gemini 3.5 Flash Lite is a text generation model from Google designed for low-cost, real-time tasks with a 1M token context window.
CohereCommand R+
Command R+ is a text generation model from Cohere designed for RAG workflows with a 128,000 token context window.
Start building with Amazon Nova Lite
No API keys required. Create AI-powered workflows with Amazon Nova Lite in minutes — free.