Nova Lite

Unknown Developer

This is a high-performance multimodal model designed for speed and affordability. It rapidly processes various data types, including images, videos, documents, and text.

Model Specifications

Technical details and capabilities of Nova Lite

Core Specifications

300.0K / 2.0K

Input / Output tokens

November 19, 2024

Release date

Capabilities & License

Multimodal Support

Supported

Web Hydrated

Yes

License

Proprietary

Resources

Research Paper

https://www.amazon.science/publications/the-amazon-nova-family-of-models-technical-report-and-model-card

API Reference

https://aws.amazon.com/bedrock/amazon-nova-lite

Performance Insights

Check out how Nova Lite handles various AI tasks through comprehensive benchmark results.

100

75

50

25

0

94.5

GSM8k

94.5

(95%)

92.4

ARC-C

92.4

(92%)

92.4

Doc VQA

92.4

(92%)

89.7

IFEval

89.7

(90%)

88.8

Translation en→Set1 COMET22

88.8

(89%)

88.8

Translation Set1→en COMET22

88.8

(89%)

86.8

Chart QA

86.8

(87%)

85.4

HumanEval

85.4

(85%)

82.4

BBH

82.4

(82%)

80.5

MMLU

80.5

(81%)

80.2

GroundUI-1K

80.2

(80%)

80.2

Text VQA

80.2

(80%)

80.2

DROP

80.2

(80%)

77.8

VATEX

77.8

(78%)

77.7

VisualWebBench

77.7

(78%)

73.6

FinQA

73.6

(74%)

73.3

MATH

73.3

(73%)

71.4

Ego Schema

71.4

(71%)

66.6

BFCL

66.6

(67%)

60.7

MM-Mind2Web

60.7

(61%)

56.2

MMMU

56.2

(56%)

43.8

CRAG

43.8

(44%)

43.1

Translation Set1→en spBleu

43.1

(43%)

42

GPQA

42

(42%)

41.5

Translation en→Set1 spBleu

41.5

(42%)

40.4

LVBench

40.4

(40%)

19.2

SQuALITY

19.2

(19%)

GSM8k

ARC-C

Doc VQA

IFEval

Translation en→Set1 COMET22

Translation Set1→en COMET22

Chart QA

HumanEval

BBH

MMLU

GroundUI-1K

Text VQA

DROP

VATEX

VisualWebBench

FinQA

MATH

Ego Schema

BFCL

MM-Mind2Web

MMMU

CRAG

Translation Set1→en spBleu

GPQA

Translation en→Set1 spBleu

LVBench

SQuALITY

Model Comparison

See how Nova Lite stacks up against other leading models across key performance metrics.

90

72

54

36

18

0

80.5

MMLU - Nova Lite

80.5

(89%)

82

MMLU - GPT-4o mini

82

(91%)

77.6

MMLU - Nova Micro

77.6

(86%)

85.9

MMLU - Nova Pro

85.9

(95%)

86.5

MMLU - GPT-4 Turbo

86.5

(96%)

87.3

MMLU - Llama 3.1 405B Instruct

87.3

(97%)

85.4

HumanEval - Nova Lite

85.4

(95%)

87.2

HumanEval - GPT-4o mini

87.2

(97%)

81.1

HumanEval - Nova Micro

81.1

(90%)

89

HumanEval - Nova Pro

89

(99%)

87.1

HumanEval - GPT-4 Turbo

87.1

(97%)

89

HumanEval - Llama 3.1 405B Instruct

89

(99%)

80.2

DROP - Nova Lite

80.2

(89%)

79.7

DROP - GPT-4o mini

79.7

(89%)

79.3

DROP - Nova Micro

79.3

(88%)

85.4

DROP - Nova Pro

85.4

(95%)

86

DROP - GPT-4 Turbo

86

(96%)

84.8

DROP - Llama 3.1 405B Instruct

84.8

(94%)

42

GPQA - Nova Lite

42

(47%)

40.2

GPQA - GPT-4o mini

40.2

(45%)

40

GPQA - Nova Micro

40

(44%)

46.9

GPQA - Nova Pro

46.9

(52%)

48

GPQA - GPT-4 Turbo

48

(53%)

50.7

GPQA - Llama 3.1 405B Instruct

50.7

(56%)

73.3

MATH - Nova Lite

73.3

(81%)

70.2

MATH - GPT-4o mini

70.2

(78%)

69.3

MATH - Nova Micro

69.3

(77%)

76.6

MATH - Nova Pro

76.6

(85%)

72.6

MATH - GPT-4 Turbo

72.6

(81%)

73.8

MATH - Llama 3.1 405B Instruct

73.8

(82%)

MMLU

HumanEval

DROP

GPQA

MATH

Nova Lite

GPT-4o mini

Nova Micro

Nova Pro

GPT-4 Turbo

Llama 3.1 405B Instruct

Detailed Benchmarks

Dive deeper into Nova Lite's performance across specific task categories. Expand each section to see detailed metrics and comparisons.

Math

GSM8k

94.5%

78.8%

70.7%

Current model

Other models

Avg (81.3%)

Coding

HumanEval

Claude 3.5 Sonnet

93.7%

87.1%

Qwen2.5 72B Instruct

86.6%

Qwen2 72B Instruct

86.0%

85.7%

85.4%

84.9%

Mistral Small 3

84.8%

Qwen2.5 7B Instruct

84.8%

Ministral 8B Instruct

34.8%

Current model

Other models

Avg (81.4%)

Reasoning

DROP

92.2%

83.4%

Claude 3.5 Haiku

83.1%

83.1%

80.9%

80.2%

79.7%

Llama 3.1 70B Instruct

79.6%

79.3%

Llama 3.1 8B Instruct

59.5%

Current model

Other models

Avg (80.1%)

Knowledge

MMLU

91.8%

82.0%

81.3%

Jamba 1.5 Large

81.2%

Mistral Small 3.1 24B

80.6%

80.5%

80.4%

Llama 3.1 Nemotron 70B Instruct

80.2%

Qwen2.5 14B Instruct

79.7%

Llama 3.2 3B Instruct

63.4%

Current model

Other models

Avg (80.1%)

GPQA

87.7%

Llama 3.2 90B Instruct

46.7%

Qwen2.5 14B Instruct

45.5%

Mistral Small 3

45.3%

Qwen2 72B Instruct

42.4%

42.0%

Llama 3.1 70B Instruct

41.7%

Claude 3.5 Haiku

41.6%

Claude 3 Sonnet

40.4%

Qwen2 7B Instruct

25.3%

Current model

Other models

Avg (45.9%)

MATH

97.9%

76.1%

Qwen2.5 7B Instruct

75.5%

74.7%

Llama 3.1 405B Instruct

73.8%

73.3%

73.0%

72.6%

Claude 3.5 Sonnet

71.1%

32.6%

Current model

Other models

Avg (72.1%)

Non categorized

ARC-C

94.8%

92.4%

90.2%

Llama 3.2 3B Instruct

78.6%

Qwen2.5 32B Instruct

70.4%

Qwen2.5 14B Instruct

67.3%

Current model

Other models

Avg (82.3%)

SQuALITY

Phi-3.5-mini-instruct

24.3%

Phi-3.5-MoE-instruct

24.1%

19.8%

19.2%

18.8%

Current model

Other models

Avg (21.2%)

LVBench

41.6%

40.4%

Current model

Other models

Avg (41.0%)

FinQA

77.2%

73.6%

65.2%

Current model

Other models

Avg (72.0%)

CRAG

Mistral Large 2

67.0%

Llama 3.3 70B Instruct

61.7%

Llama 3.1 8B Instruct

60.0%

Ministral 8B Instruct

52.0%

50.3%

49.4%

43.8%

43.1%

Current model

Other models

Avg (53.4%)

VisualWebBench

79.7%

77.7%

Current model

Other models

Avg (78.7%)

MM-Mind2Web

63.7%

60.7%

Current model

Other models

Avg (62.2%)

GroundUI-1K

81.4%

80.2%

Current model

Other models

Avg (80.8%)

BFCL

Llama 3.1 405B Instruct

88.5%

Llama 3.1 70B Instruct

84.8%

Llama 3.1 8B Instruct

76.1%

74.0%

68.4%

66.6%

66.4%

65.0%

60.0%

56.2%

Current model

Other models

Avg (70.6%)

MMMU

Gemini Pro 2.5 Experimental

81.7%

Gemini 1.5 Flash

62.3%

61.7%

Llama 3.2 90B Instruct

60.3%

59.4%

56.2%

Gemini 1.5 Flash 8B

53.7%

53.6%

53.6%

0.0%

Current model

Other models

Avg (54.2%)

Chart QA

89.2%

86.8%

Current model

Other models

Avg (88.0%)

Doc VQA

93.5%

92.4%

Current model

Other models

Avg (93.0%)

Text VQA

81.5%

80.2%

Llama 3.2 90B Instruct

73.5%

Current model

Other models

Avg (78.4%)

VATEX

77.8%

77.8%

Current model

Other models

Avg (77.8%)

Ego Schema

72.1%

71.4%

Current model

Other models

Avg (71.8%)

Translation en→Set1 spBleu

43.4%

41.5%

40.2%

Current model

Other models

Avg (41.7%)

Translation en→Set1 COMET22

89.1%

88.8%

88.5%

Current model

Other models

Avg (88.8%)

Translation Set1→en spBleu

44.4%

43.1%

42.6%

Current model

Other models

Avg (43.4%)

Translation Set1→en COMET22

89.0%

88.8%

88.7%

Current model

Other models

Avg (88.8%)

IFEval

Claude 3.7 Sonnet

93.2%

92.1%

Llama 3.3 70B Instruct

92.1%

90.4%

90.0%

89.7%

Llama 3.1 405B Instruct

88.6%

Llama 3.1 70B Instruct

87.5%

87.2%

61.3%

Current model

Other models

Avg (87.2%)

BBH

87.6%

86.9%

Qwen2.5 32B Instruct

84.5%

84.3%

82.4%

Qwen2 72B Instruct

82.4%

79.5%

Qwen2.5 14B Instruct

78.2%

70.6%

Current model

Other models

Avg (81.8%)

Providers Pricing Coming Soon

We're working on gathering comprehensive pricing data from all major providers for Nova Lite. Compare costs across platforms to find the best pricing for your use case.

OpenAI

Anthropic

Google

Mistral AI

Cohere

Share your feedback

Hi, I'm Charlie Palars, the founder of Deepranking.ai. I'm always looking for ways to improve the site and make it more useful for you. You can write me through this form or directly through X at @palarsio.

Your feedback helps us improve our service

Stay Ahead with AI Updates

Get insights on Gemini Pro 2.5, Sonnet 3.7 and more top AI models