Ministral 8B Instruct

Mistral AI

The Ministral-8B-Instruct-2410 is a fine-tuned instruction-following model designed for superior local and on-device intelligence, as well as edge computing applications. It demonstrates a marked performance increase compared to other models with similar parameters.

Model Specifications

Technical details and capabilities of Ministral 8B Instruct

Core Specifications

8.0B Parameters

Model size and complexity

128.0K / 128.0K

Input / Output tokens

October 15, 2024

Release date

Capabilities & License

Multimodal Support

Not Supported

Web Hydrated

No

License

Mistral Research License

Resources

API Reference

https://huggingface.co/mistralai/Ministral-8B-Instruct-2410

Performance Insights

Check out how Ministral 8B Instruct handles various AI tasks through comprehensive benchmark results.

90

68

45

23

0

83

MT-Bench

83

(92%)

75.3

Winogrande

75.3

(84%)

71.9

ARC Challenge

71.9

(80%)

70.9

Arena Hard

70.9

(79%)

70

MBPP pass@1

70

(78%)

65.5

TriviaQA

65.5

(73%)

65

MMLU

65

(72%)

57.5

French MMLU

57.5

(64%)

54.5

MATH

54.5

(61%)

52

CRAG

52

(58%)

48.3

AGI Eval

48.3

(54%)

34.8

HumanEval

34.8

(39%)

33

HELMET LongQA

33

(37%)

19.2

FinanceBench

19.2

(21%)

17.5

LongBench

17.5

(19%)

MT-Bench

Winogrande

ARC Challenge

Arena Hard

MBPP pass@1

TriviaQA

MMLU

French MMLU

MATH

CRAG

AGI Eval

HumanEval

HELMET LongQA

FinanceBench

LongBench

Model Comparison

See how Ministral 8B Instruct stacks up against other leading models across key performance metrics.

80

64

48

32

16

0

70.9

Arena Hard - Ministral 8B Instruct

70.9

(89%)

65.8

Arena Hard - Llama 3.3 70B Instruct

65.8

(82%)

33.1

Arena Hard - Command R+

33.1

(41%)

28.2

Arena Hard - Llama 3.1 8B Instruct

28.2

(35%)

51.2

Arena Hard - Jamba 1.6 Mini

51.2

(64%)

76.5

Arena Hard - Jamba 1.6 Large

76.5

(96%)

52

CRAG - Ministral 8B Instruct

52

(65%)

61.7

CRAG - Llama 3.3 70B Instruct

61.7

(77%)

49.4

CRAG - Command R+

49.4

(62%)

60

CRAG - Llama 3.1 8B Instruct

60

(75%)

76.2

CRAG - Jamba 1.6 Mini

76.2

(95%)

78.2

CRAG - Jamba 1.6 Large

78.2

(98%)

19.2

FinanceBench - Ministral 8B Instruct

19.2

(24%)

20

FinanceBench - Llama 3.3 70B Instruct

20

(25%)

13.3

FinanceBench - Command R+

13.3

(17%)

28.4

FinanceBench - Llama 3.1 8B Instruct

28.4

(36%)

45.4

FinanceBench - Jamba 1.6 Mini

45.4

(57%)

64.5

FinanceBench - Jamba 1.6 Large

64.5

(81%)

33

HELMET LongQA - Ministral 8B Instruct

33

(41%)

52.8

HELMET LongQA - Llama 3.3 70B Instruct

52.8

(66%)

44.8

HELMET LongQA - Command R+

44.8

(56%)

29.2

HELMET LongQA - Llama 3.1 8B Instruct

29.2

(37%)

46.9

HELMET LongQA - Jamba 1.6 Mini

46.9

(59%)

56.7

HELMET LongQA - Jamba 1.6 Large

56.7

(71%)

17.5

LongBench - Ministral 8B Instruct

17.5

(22%)

21.7

LongBench - Llama 3.3 70B Instruct

21.7

(27%)

18.9

LongBench - Command R+

18.9

(24%)

17.7

LongBench - Llama 3.1 8B Instruct

17.7

(22%)

32

LongBench - Jamba 1.6 Mini

32

(40%)

38.8

LongBench - Jamba 1.6 Large

38.8

(49%)

Arena Hard

CRAG

FinanceBench

HELMET LongQA

LongBench

Ministral 8B Instruct

Llama 3.3 70B Instruct

Command R+

Llama 3.1 8B Instruct

Jamba 1.6 Mini

Jamba 1.6 Large

Detailed Benchmarks

Dive deeper into Ministral 8B Instruct's performance across specific task categories. Expand each section to see detailed metrics and comparisons.

Coding

HumanEval

72.0%

Phi-3.5-MoE-instruct

70.7%

68.0%

67.0%

Phi-3.5-mini-instruct

62.8%

51.8%

40.2%

Ministral 8B Instruct

34.8%

Current model

Other models

Avg (58.4%)

Knowledge

MMLU

69.7%

Llama 3.1 8B Instruct

69.4%

69.2%

Phi-3.5-mini-instruct

69.0%

Mistral NeMo Instruct

68.0%

Qwen2.5-Coder 7B Instruct

67.6%

Ministral 8B Instruct

65.0%

Llama 3.2 3B Instruct

63.4%

Current model

Other models

Avg (67.7%)

MATH

97.9%

Qwen2 72B Instruct

59.7%

Phi-3.5-MoE-instruct

59.5%

Gemini 1.5 Flash 8B

58.7%

Qwen2.5-Coder 32B Instruct

57.2%

Ministral 8B Instruct

54.5%

Llama 3.2 11B Instruct

51.9%

50.6%

49.7%

32.6%

Current model

Other models

Avg (57.2%)

Non categorized

Winogrande

87.5%

85.4%

Qwen2 72B Instruct

85.1%

Llama 3.1 Nemotron 70B Instruct

84.5%

Qwen2.5 32B Instruct

82.0%

80.6%

Mistral NeMo Instruct

76.8%

Ministral 8B Instruct

75.3%

Current model

Other models

Avg (82.2%)

MT-Bench

Llama 3.1 Nemotron 70B Instruct

9.0

90.2%

Mistral Large 2

86.3%

Qwen2 7B Instruct

84.1%

Mistral Small 3

83.5%

Ministral 8B Instruct

83.0%

76.8%

Current model

Other models

Avg (2.0)

Arena Hard

Mistral Small 3

87.6%

Jamba 1.6 Large

76.5%

Mistral Large 2

71.5%

Ministral 8B Instruct

70.9%

Llama 3.3 70B Instruct

65.8%

Jamba 1.5 Large

65.4%

51.2%

46.1%

Llama 3.1 8B Instruct

28.2%

Current model

Other models

Avg (62.6%)

ARC Challenge

Llama 3.1 405B Instruct

96.9%

89.2%

85.7%

Phi-3.5-mini-instruct

84.6%

Llama 3.1 8B Instruct

83.4%

Ministral 8B Instruct

71.9%

71.0%

Llama 3.1 Nemotron 70B Instruct

69.2%

Qwen2 72B Instruct

68.9%

Current model

Other models

Avg (80.1%)

TriviaQA

88.0%

83.7%

Mistral Small 3.1 24B

80.5%

76.6%

Mistral NeMo Instruct

73.8%

Ministral 8B Instruct

65.5%

Current model

Other models

Avg (78.0%)

CRAG

Jamba 1.6 Large

78.2%

76.2%

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 (58.2%)

FinanceBench

Jamba 1.6 Large

64.5%

45.4%

Llama 3.1 8B Instruct

28.4%

Llama 3.3 70B Instruct

20.0%

Ministral 8B Instruct

19.2%

13.3%

Current model

Other models

Avg (31.8%)

HELMET LongQA

Jamba 1.6 Large

56.7%

Llama 3.3 70B Instruct

52.8%

46.9%

44.8%

Ministral 8B Instruct

33.0%

Llama 3.1 8B Instruct

29.2%

Mistral Large 2

26.7%

Current model

Other models

Avg (41.4%)

LongBench

Jamba 1.6 Large

38.8%

Mistral Large 2

34.4%

32.0%

Llama 3.3 70B Instruct

21.7%

18.9%

Llama 3.1 8B Instruct

17.7%

Ministral 8B Instruct

17.5%

Current model

Other models

Avg (25.9%)

Providers Pricing Coming Soon

We're working on gathering comprehensive pricing data from all major providers for Ministral 8B Instruct. 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