Amazon Nova Lite 1.0

Amazon's new lower cost model. Can handle up to 300k input tokens, with faster output but less thorough understanding than Amazon's Nova Pro.

Added Dec 3, 2024

Pricing

Auto routing · per 1M tokens

Input: $0.060
Output: $0.24

Compare provider prices

Specifications

Context window: 300K
Max output: 5.1K
Avg output (7d): 759 tokens; Longer than 60% of models

Benchmarks

Artificial Analysis

LMArena

Vectara

Sourced from Artificial Analysis.

Intelligence Index

6.7

Better than 19% of models compared

Agentic work

T²-Bench Telecom (legacy)

Legacy fallback · Conversational AI agents in dual-control scenarios

17.5%

Better than 16% of models compared

Document reasoning

AA-LCR v1.1

Long context reasoning with updated grading

19.0%

Better than 21% of models compared

MLCR-AA

Medical long-context reasoning

0.0%

Better than 1% of models compared

Reasoning

HLE

Humanity's Last Exam

4.3%

Better than 18% of models compared

IFBench

Instruction-following benchmark

34.1%

Better than 24% of models compared

CritPt

Research-level physics reasoning

0.0%

Coding

Terminal-Bench Hard (legacy)

Legacy fallback · Agentic coding and terminal use

0.8%

Better than 13% of models compared

LiveCodeBench

Contamination-free coding benchmark

16.7%

Better than 17% of models compared

Math

AIME 2025

American Invitational Mathematics Examination 2025

7.0%

Better than 11% of models compared

AIME

American Invitational Mathematics Examination

10.7%

Better than 34% of models compared

Math-500

Diverse mathematical problem solving benchmark

76.5%

Better than 39% of models compared

Knowledge

MMLU-Pro

Professional and academic subject knowledge

59.0%

Better than 22% of models compared

AA-Omniscience Accuracy

Proportion of correctly answered questions

9.8%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

57.6%

Legacy benchmarks

GPQA Diamond (legacy)

Graduate-level scientific reasoning

43.3%

Better than 18% of models compared

AA-LCR (unversioned / legacy)

Long context reasoning evaluation