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.
Aggregate Rating 1510 ± 15, rank #754 of 1656 rated models, from 192 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MLCR-AA - Conciseness | 100 | Conciseness (%) | 99.5 |
| HELM NaturalQuestions (Open) | 81.52 | F1 (%) | 98.9 |
| HELM Lite - LegalBench - International Citizenship Questions | 65.8 | Exact Match (%) | 96.7 |
| Ko-AgentBench - L6 Efficient Tool Utilization | 31.33 | Efficiency Score (%) | 92.9 |
| HELM MMLU - High School Geography | 94.44 | EM | 87.8 |
| HELM Lite - MATH Level 1 - Number Theory | 86.67 | Equivalent (CoT) (%) | 87.2 |
| HELM Lite - MATH Level 1 - Precalculus | 78.95 | Equivalent (CoT) (%) | 86.1 |
| HELM MMLU - High School Mathematics | 58.15 | EM | 84.6 |
| HELM NarrativeQA | 76.81 | F1 (%) | 81.1 |
| HELM MMLU - High School Physics | 68.87 | EM | 80.1 |
| HELM MMLU - College Physics | 60.78 | EM | 78.2 |
| HELM MMLU - Elementary Mathematics | 75.66 | EM | 78.2 |
Interactive version: theaggregate.ai/model?slug=nova-lite · How It Works · Data refreshed daily, snapshot 2026-10-20.