わたしたちの信念

AA Intelligence Index

このベンチマークについて

Artificial Analysisが算出する総合知能指数。推論・コーディング・数学・言語理解など複数のベンチマークを加重平均したものです。

出典: Artificial Analysis
順位モデル
#1

Anthropic

Claude Opus 4.7

57.3
#2

Google

Gemini 3.1 Pro

57.2
#3

OpenAI

GPT-5.4

56.8
#4

Anthropic

Claude Opus 4.6

53.0
#5

Meta

Muse Spark

52.1
#6

DeepSeek

DeepSeek V4 Pro

51.5
#7

Z.ai

GLM-5.1

51.4
#8

Alibaba

Qwen3.6 Plus

50.0
#9

Z.ai

GLM-5

49.8
#10

Anthropic

Claude Opus 4.5

49.7
#11

MiniMax

MiniMax M2.7

49.6
#12

Grok

Grok 4.20 (Reasoning)

49.3
#13

Xiaomi

MiMo-V2-Pro

49.2
#14

Moonshot AI

Kimi K2.5

46.8
#15

DeepSeek

DeepSeek V4 Flash

46.5
#16

Google

Gemini 3 Flash

46.4
#17

Anthropic

Claude Sonnet 4.5

43.0
#18

Anthropic

Claude Sonnet 4.6

42.6
#19

Anthropic

Claude Opus 4.1

42.0
#20

MiniMax

MiniMax M2.5

41.9
#21

DeepSeek

DeepSeek V3.2

41.7
#22

Alibaba

Qwen3.5 397B A17B

40.1
#23

Google

Gemma 4 31B

39.2
#24

Anthropic

Claude Opus 4

39.0
#25

Anthropic

Claude Sonnet 4

38.7
#26

Grok

Grok 4.1 Fast (Reasoning)

38.6
#27

OpenAI

GPT-5.4 Nano

38.1
#28

OpenAI

GPT-5.4 Mini

37.7
#29

Anthropic

Claude Haiku 4.5

37.1
#30

NVIDIA

Nemotron 3 Super

36.0
#31

Google

Gemini 2.5 Pro

34.6
#32

Google

Gemini 3.1 Flash Lite

33.5
#33

LG AI Research

K-EXAONE

32.0
#34

Arcee AI

Trinity Large Thinking

31.9
#35

Baidu

ERNIE 5.0 Thinking

29.1
#36

Grok

Grok 4.20

29.0
#37

Mistral AI

Mistral Small 4

27.8
#38

Google

Gemini 2.5 Flash

27.0
#39

OpenAI

GPT-4.1

26.3
#40

OpenAI

GPT-5 Nano

25.9
#41

OpenAI

GPT OSS 120B

24.5
#42

Meituan

Longcat Flash Chat

23.9
#43

Grok

Grok 4.1 Fast

23.6
#44

OpenAI

GPT-5

21.8
#45

OpenAI

GPT-5 Mini

20.7
#46

Google

Gemini 2.5 Flash Lite

19.4
#47

Meta

Llama 4 Maverick

18.4
#48

Amazon

Nova 2 Lite

18.0
#49

Baidu

ERNIE 4.5 300B A47B

15.0
#50

Meta

Llama 4 Scout

13.5