GPT-4o mini

OpenAI

GPT-4o mini is OpenAI's new, budget-friendly model intended to broaden access to AI. Surpassing earlier models like GPT-3.5 Turbo, it demonstrates exceptional text comprehension and multimodal reasoning. It supports both text and vision within a large 128K token context window, enabling real-time, cost-effective solutions such as customer service chatbots. Its pricing is a significant improvement over previous models, at just 15 cents per million input tokens and 60 cents per million output tokens. Furthermore, GPT-4o mini includes robust safety features and enhanced defenses against security risks.

Model Specifications

Technical details and capabilities of GPT-4o mini

Core Specifications

128.0K / 16.4K

Input / Output tokens

September 30, 2023

Knowledge cutoff date

July 17, 2024

Release date

Capabilities & License

Multimodal Support

Supported

Web Hydrated

License

Proprietary

Resources

API Reference

https://platform.openai.com/docs/api-reference

Performance Insights

Check out how GPT-4o mini handles various AI tasks through comprehensive benchmark results.

87.2

HumanEval

87.2

(97%)

MGSM

(97%)

MMLU

(91%)

79.7

DROP

79.7

(89%)

70.2

MATH

70.2

(78%)

59.4

MMMU

59.4

(66%)

56.7

MathVista

56.7

(63%)

40.2

GPQA

40.2

(45%)

HumanEval

MGSM

MMLU

DROP

MATH

MMMU

MathVista

GPQA

Model Comparison

See how GPT-4o mini stacks up against other leading models across key performance metrics.

100

MMLU - GPT-4o mini

(82%)

86.5

MMLU - GPT-4 Turbo

86.5

(87%)

MMLU - Llama 3.3 70B Instruct

(86%)

86.8

MMLU - Claude 3 Opus

86.8

(87%)

88.7

MMLU - GPT-4o

88.7

(89%)

78.9

MMLU - Gemini 1.5 Flash

78.9

(79%)

87.2

HumanEval - GPT-4o mini

87.2

(87%)

87.1

HumanEval - GPT-4 Turbo

87.1

(87%)

88.4

HumanEval - Llama 3.3 70B Instruct

88.4

(88%)

84.9

HumanEval - Claude 3 Opus

84.9

(85%)

90.2

HumanEval - GPT-4o

90.2

(90%)

74.3

HumanEval - Gemini 1.5 Flash

74.3

(74%)

40.2

GPQA - GPT-4o mini

40.2

(40%)

GPQA - GPT-4 Turbo

(48%)

50.5

GPQA - Llama 3.3 70B Instruct

50.5

(51%)

50.4

GPQA - Claude 3 Opus

50.4

(50%)

53.6

GPQA - GPT-4o

53.6

(54%)

GPQA - Gemini 1.5 Flash

(51%)

MGSM - GPT-4o mini

(87%)

88.5

MGSM - GPT-4 Turbo

88.5

(89%)

91.1

MGSM - Llama 3.3 70B Instruct

91.1

(91%)

90.7

MGSM - Claude 3 Opus

90.7

(91%)

90.5

MGSM - GPT-4o

90.5

(91%)

82.6

MGSM - Gemini 1.5 Flash

82.6

(83%)

70.2

MATH - GPT-4o mini

70.2

(70%)

72.6

MATH - GPT-4 Turbo

72.6

(73%)

MATH - Llama 3.3 70B Instruct

(77%)

60.1

MATH - Claude 3 Opus

60.1

(60%)

76.6

MATH - GPT-4o

76.6

(77%)

77.9

MATH - Gemini 1.5 Flash

77.9

(78%)

MMLU

HumanEval

GPQA

MGSM

MATH

GPT-4o mini

GPT-4 Turbo

Llama 3.3 70B Instruct

Claude 3 Opus

GPT-4o

Gemini 1.5 Flash

Detailed Benchmarks

Dive deeper into GPT-4o mini's performance across specific task categories. Expand each section to see detailed metrics and comparisons.

Coding

HumanEval

93.7%

88.4%

88.1%

87.8%

87.2%

87.1%

86.6%

86.0%

Ministral 8B Instruct

34.8%

Current model

Other models

Avg (82.8%)

Reasoning

DROP

92.2%

83.1%

83.1%

80.9%

80.2%

79.7%

Llama 3.1 70B Instruct

79.6%

Nova Micro

79.3%

Claude 3 Sonnet

78.9%

Llama 3.1 8B Instruct

59.5%

Current model

Other models

Avg (79.7%)

Knowledge

MMLU

91.8%

Command A

84.0%

Llama 3.1 70B Instruct

83.6%

83.3%

82.3%

82.0%

81.3%

81.2%

Mistral Small 3.1 24B

80.6%

Llama 3.2 3B Instruct

63.4%

Current model

Other models

Avg (81.4%)

GPQA

87.7%

Nova Lite

42.0%

Llama 3.1 70B Instruct

41.7%

41.6%

40.4%

40.2%

40.0%

38.4%

36.9%

25.3%

Current model

Other models

Avg (43.4%)

MATH

97.9%

73.0%

72.6%

71.1%

70.6%

70.2%

69.4%

69.3%

Mistral Small 3.1 24B

69.3%

Gemini 1.0 Pro

32.6%

Current model

Other models

Avg (69.6%)

Non categorized

MMMU

Gemini Pro 2.5 Experimental

81.7%

Mistral Small 3.1 24B

62.8%

Gemini 1.5 Flash

62.3%

Nova Pro

61.7%

Llama 3.2 90B Instruct

60.3%

59.4%

56.2%

53.7%

53.6%

0.0%

Current model

Other models

Avg (55.2%)

MGSM

o3-mini

92.0%

GPT-4o

90.5%

89.3%

GPT-4 Turbo

88.5%

Gemini 1.5 Pro

87.5%

GPT-4o mini

87.0%

Llama 3.2 90B Instruct

86.9%

Claude 3.5 Haiku

85.6%

Claude 3 Sonnet

83.5%

Phi-3.5-mini-instruct

47.9%

Current model

Other models

Avg (83.9%)

MathVista

74.9%

65.8%

63.8%

63.8%

Llama 3.2 90B Instruct

57.3%

GPT-4o mini

56.7%

Gemini 1.5 Flash 8B

54.7%

Grok-1.5

52.8%

Llama 3.2 11B Instruct

51.5%

GPT-3.5 Turbo

0.0%

Current model

Other models

Avg (54.1%)

Providers Pricing Coming Soon

We're working on gathering comprehensive pricing data from all major providers for GPT-4o mini. 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

GPT-4o mini

Model Specifications

Core Specifications

Capabilities & License

Resources

Performance Insights

Model Comparison

Detailed Benchmarks

Coding

HumanEval

Reasoning

DROP

Knowledge

MMLU

GPQA

MATH

Non categorized

MMMU

MGSM

MathVista

Providers Pricing Coming Soon

Share your feedback

Stay Ahead with AI Updates