우리들의 비전

Arena 창작 Elo

이 벤치마크는?

창작 글쓰기 프롬프트에서의 사용자 선호도를 집계한 Arena Elo 점수입니다.

출처: Arena Intelligence
순위모델
#1

Anthropic

Claude Opus 4.7

1499
#2

Anthropic

Claude Opus 4.6

1494
#3

Google

Gemini 3.1 Pro

1488
#4

Anthropic

Claude Opus 4.5

1468
#5

Grok

Grok 4.20

1467
#6

Grok

Grok 4.20 (Reasoning)

1467
#7

Meta

Muse Spark

1463
#8

Google

Gemini 3 Flash

1460
#9

Z.ai

GLM-5.1

1451
#10

OpenAI

GPT-5.4

1448
#11

OpenAI

GPT-5.4 Pro

1448
#12

DeepSeek

DeepSeek V4 Pro

1448
#13

Z.ai

GLM-5

1447
#14

Google

Gemini 2.5 Pro

1447
#15

Anthropic

Claude Opus 4.1

1445
#16

Anthropic

Claude Sonnet 4.6

1444
#17

Anthropic

Claude Sonnet 4.5

1442
#18

Anthropic

Claude Opus 4

1429
#19

Baidu

ERNIE 5.0 Thinking

1425
#20

Moonshot AI

Kimi K2.6

1422
#21

Google

Gemma 4 31B

1422
#22

Xiaomi

MiMo-V2-Pro

1421
#23

Google

Gemini 3.1 Flash Lite

1419
#24

OpenAI

GPT-5.4 Mini

1417
#25

Moonshot AI

Kimi K2.5

1416
#26

Alibaba

Qwen3.5 397B A17B

1415
#27

Grok

Grok 4.1 Fast

1412
#28

Grok

Grok 4.1 Fast (Reasoning)

1412
#29

DeepSeek

DeepSeek V4 Flash

1404
#30

DeepSeek

DeepSeek V3.2

1399
#31

Google

Gemini 2.5 Flash

1398
#32

Anthropic

Claude Sonnet 4

1395
#33

Meituan

Longcat Flash Chat

1390
#34

Alibaba

Qwen3.6 Plus

1390
#35

Anthropic

Claude Haiku 4.5

1385
#36

MiniMax

MiniMax M2.5

1376
#37

OpenAI

GPT-5

1375
#38

Google

Gemini 2.5 Flash Lite

1361
#39

Arcee AI

Trinity Large Thinking

1357
#40

MiniMax

MiniMax M2.7

1350
#41

OpenAI

GPT-5.4 Nano

1336
#42

OpenAI

GPT-5 Mini

1325
#43

Meta

Llama 4 Maverick

1307
#44

NVIDIA

Nemotron 3 Super

1301
#45

Meta

Llama 4 Scout

1290
#46

OpenAI

GPT-4.1

1285
#47

OpenAI

GPT OSS 120B

1279
#48

Amazon

Nova 2 Lite

1276
#49

OpenAI

GPT-5 Nano

1248