Nova Lite: benchmark results

Amazon's low-cost multimodal Bedrock model (December 2024) taking text, images, and video in a 300K context. Provider: Amazon. Released 2024-12-03. Access: API.

Unified ELO 1480 ± 1, rank #786 of 1392 rated models, from 86 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
HELM NaturalQuestions (Open)81.52F1 (%)98.9
Ko-AgentBench - L6 Efficient Tool Utilization31.33Efficiency Score (%)92.9
HELM NarrativeQA76.81F1 (%)81.1
HELM (Stanford)70.8Mean Win Rate (%)77.8
Translation en to Set1 COMET2288.8en→Set1 COMET22 (self-reported)66.7
Translation en to Set1 spBleu41.5en→Set1 spBLEU (self-reported)66.7
HELM WMT 201420.43BLEU-4 (%)65
LLM Stats (DROP)80.2Score (%)64.3
LLM Stats (ChartQA)86.8Score (%)64
Translation Set1 to en COMET2288.8Set1→en COMET22 (self-reported)62.5
Translation Set1 to en spBleu43.1Set1→en spBLEU (self-reported)58.3
AA Omniscience - Software Engineering (SWE) - Julia8.33Accuracy (%)52.6

Interactive version: theaggregate.ai/model?slug=nova-lite · How It Works · Data refreshed daily, snapshot 2026-09-05.