우리들의 비전

생활·사회과학

이 벤치마크는?

과학 관련 전문가 프롬프트에서의 Arena Elo 점수입니다.

출처: Arena Intelligence
순위모델
#1

Anthropic

Claude Opus 4.7

1521
#2

Anthropic

Claude Opus 4.6

1517
#3

Google

Gemini 3.1 Pro

1513
#4

Meta

Muse Spark

1500
#5

Anthropic

Claude Sonnet 4.6

1494
#6

Z.ai

GLM-5.1

1490
#7

Anthropic

Claude Opus 4.5

1489
#8

Grok

Grok 4.20

1487
#9

Grok

Grok 4.20 (Reasoning)

1487
#10

Google

Gemini 3 Flash

1486
#11

Z.ai

GLM-5

1482
#12

OpenAI

GPT-5.4

1479
#13

OpenAI

GPT-5.4 Pro

1479
#14

Xiaomi

MiMo-V2-Pro

1479
#15

DeepSeek

DeepSeek V4 Pro

1478
#16

Google

Gemini 2.5 Pro

1473
#17

Moonshot AI

Kimi K2.6

1472
#18

Anthropic

Claude Opus 4.1

1471
#19

Moonshot AI

Kimi K2.5

1471
#20

Anthropic

Claude Sonnet 4.5

1467
#21

OpenAI

GPT-5.4 Mini

1465
#22

Google

Gemma 4 31B

1464
#23

Alibaba

Qwen3.5 397B A17B

1463
#24

Google

Gemini 3.1 Flash Lite

1462
#25

Baidu

ERNIE 5.0 Thinking

1461
#26

DeepSeek

DeepSeek V4 Flash

1460
#27

Meituan

Longcat Flash Chat

1453
#28

Grok

Grok 4.1 Fast

1450
#29

Grok

Grok 4.1 Fast (Reasoning)

1450
#30

DeepSeek

DeepSeek V3.2

1448
#31

Alibaba

Qwen3.6 Plus

1447
#32

OpenAI

GPT-5

1443
#33

Anthropic

Claude Opus 4

1439
#34

Google

Gemini 2.5 Flash

1430
#35

Anthropic

Claude Haiku 4.5

1422
#36

OpenAI

GPT-5.4 Nano

1421
#37

Anthropic

Claude Sonnet 4

1418
#38

MiniMax

MiniMax M2.5

1408
#39

MiniMax

MiniMax M2.7

1407
#40

Google

Gemini 2.5 Flash Lite

1403
#41

Arcee AI

Trinity Large Thinking

1402
#42

OpenAI

GPT-5 Mini

1396
#43

NVIDIA

Nemotron 3 Super

1366
#44

OpenAI

GPT OSS 120B

1361
#45

Amazon

Nova 2 Lite

1356
#46

OpenAI

GPT-5 Nano

1345
#47

Meta

Llama 4 Maverick

1344
#48

Meta

Llama 4 Scout

1336
#49

OpenAI

GPT-4.1

1321