Nova Lite
Multimodal
Zero-eval
#1VATEX
#2Translation Set1→en COMET22
#2Translation en→Set1 COMET22
+7 more
by Amazon
About
Nova Lite is a multimodal language model developed by Amazon. It achieves strong performance with an average score of 70.7% across 27 benchmarks. It excels particularly in GSM8k (94.5%), DocVQA (92.4%), ARC-C (92.4%). It supports a 302K token context window for handling large documents. The model is available through 1 API provider. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2024, it represents Amazon's latest advancement in AI technology.
Pricing Range
Input (per 1M)$0.06 -$0.06
Output (per 1M)$0.24 -$0.24
Providers1
Timeline
AnnouncedNov 20, 2024
ReleasedNov 20, 2024
Specifications
Capabilities
Multimodal
License & Family
License
Proprietary
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
27 benchmarks
Average Score
70.7%
Best Score
94.5%
High Performers (80%+)
13Performance Metrics
Max Context Window
302.0KAvg Throughput
100.0 tok/sAvg Latency
1msTop Categories
reasoning
92.4%
code
87.5%
math
83.9%
vision
76.6%
general
64.9%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
GSM8k
Rank #14 of 46
#11Qwen2.5 14B Instruct
94.8%
#12Nova Pro
94.8%
#13Claude 3 Opus
95.0%
#14Nova Lite
94.5%
#15Gemma 3 12B
94.4%
#16Qwen3 235B A22B
94.4%
#17Mistral Large 2
93.0%
DocVQA
Rank #16 of 26
#13GPT-4o
92.8%
#14Phi-4-multimodal-instruct
93.2%
#15Grok-2 mini
93.2%
#16Nova Lite
92.4%
#17DeepSeek VL2 Small
92.3%
#18Pixtral-12B
90.7%
#19Llama 3.2 90B Instruct
90.1%
ARC-C
Rank #7 of 31
#4Jamba 1.5 Large
93.0%
#5Claude 3 Sonnet
93.2%
#6Llama 3.1 70B Instruct
94.8%
#7Nova Lite
92.4%
#8Mistral Small 3 24B Base
91.3%
#9Phi-3.5-MoE-instruct
91.0%
#10Nova Micro
90.2%
IFEval
Rank #8 of 37
#5Kimi K2 Instruct
89.8%
#6Gemma 3 4B
90.2%
#7Gemma 3 27B
90.4%
#8Nova Lite
89.7%
#9Llama 3.1 Nemotron Ultra 253B v1
89.5%
#10Gemma 3 12B
88.9%
#11Qwen3-235B-A22B-Instruct-2507
88.7%
Translation Set1→en COMET22
Rank #2 of 3
#1Nova Pro
89.0%
#2Nova Lite
88.8%
#3Nova Micro
88.7%
All Benchmark Results for Nova Lite
Complete list of benchmark scores with detailed information
GSM8k GSM8k benchmark | math | text | 0.94 | 94.5% | Self-reported |
DocVQA DocVQA benchmark | vision | multimodal | 0.92 | 92.4% | Self-reported |
ARC-C ARC-C benchmark | reasoning | text | 0.92 | 92.4% | Self-reported |
IFEval IFEval benchmark | code | text | 0.90 | 89.7% | Self-reported |
Translation Set1→en COMET22 Translation Set1→en COMET22 benchmark | general | text | 0.89 | 88.8% | Self-reported |
Translation en→Set1 COMET22 Translation en→Set1 COMET22 benchmark | general | text | 0.89 | 88.8% | Self-reported |
ChartQA ChartQA benchmark | general | multimodal | 0.87 | 86.8% | Self-reported |
HumanEval HumanEval benchmark | code | text | 0.85 | 85.4% | Self-reported |
BBH BBH benchmark | general | text | 0.82 | 82.4% | Self-reported |
MMLU MMLU benchmark | general | text | 0.81 | 80.5% | Self-reported |
Showing 1 to 10 of 27 benchmarks
Resources