Arena 코딩 Elo

이 벤치마크는?

코딩 관련 프롬프트에서의 사용자 선호도를 집계한 Arena Elo 점수입니다.

출처: Arena Intelligence
순위모델
#1

Anthropic

Claude Opus 4.7

1573
#2

Anthropic

Claude Opus 4.6

1554
#3

Anthropic

Claude Opus 4.5

1531
#4

Google

Gemini 3.1 Pro

1529
#5

Meta

Muse Spark

1529
#6

OpenAI

GPT-5.4

1527
#7

OpenAI

GPT-5.4 Pro

1527
#8

Z.ai

GLM-5.1

1524
#9

Anthropic

Claude Sonnet 4.6

1521
#10

Anthropic

Claude Sonnet 4.5

1520
#11

Moonshot AI

Kimi K2.6

1516
#12

ByteDance

Dola Seed 2.0 Pro

1514
#13

Anthropic

Claude Opus 4.1

1513
#14

xAI

Grok 4.20

1511
#15

xAI

Grok 4.20 (Reasoning)

1511
#16

Xiaomi

MiMo V2.5 Pro

1511
#17

Google

Gemini 3 Flash

1509
#18

OpenAI

GPT-5.5

1509
#19

OpenAI

GPT-5.4 Mini

1508
#20

Moonshot AI

Kimi K2.5

1507
#21

Xiaomi

MiMo V2 Pro

1506
#22

Alibaba

Qwen3.6 Plus

1506
#23

Google

Gemma 4 31B

1498
#24

Anthropic

Claude Opus 4

1498
#25

Alibaba

Qwen3.6 Max

1498
#26

Meituan

Longcat Flash Chat

1497
#27

xAI

Grok 4.3

1495
#28

Baidu

ERNIE 5.0 Thinking

1494
#29

Z.ai

GLM-5

1493
#30

Xiaomi

MiMo V2.5

1489
#31

Alibaba

Qwen3.5 397B A17B

1488
#32

DeepSeek

DeepSeek V4 Flash

1480
#33

DeepSeek

DeepSeek V4 Pro

1480
#34

Anthropic

Claude Haiku 4.5

1478
#35

Anthropic

Claude Sonnet 4

1473
#36

Tencent

Hy3

1471
#37

DeepSeek

DeepSeek V3.2

1468
#38

OpenAI

GPT-5.4 Nano

1467
#39

OpenAI

GPT-5

1466
#40

MiniMax

MiniMax M2.7

1466
#41

Google

Gemini 2.5 Pro

1465
#42

xAI

Grok 4.1 Fast

1465
#43

xAI

Grok 4.1 Fast (Reasoning)

1465
#44

Google

Gemini 3.1 Flash Lite

1461
#45

MiniMax

MiniMax M2.5

1454
#46

Arcee AI

Trinity Large Thinking

1443
#47

OpenAI

GPT-5 Mini

1429
#48

Google

Gemini 2.5 Flash

1424
#49

NVIDIA

Nemotron 3 Super

1409
#50

Google

Gemini 2.5 Flash Lite

1398
#51

Amazon

Nova 2 Lite

1397
#52

OpenAI

GPT OSS 120B

1390
#53

OpenAI

GPT-5 Nano

1381
#54

Meta

Llama 4 Maverick

1373
#55

NVIDIA

Nemotron 3 Nano 30B A3B

1364
#56

Meta

Llama 4 Scout

1361
#57

OpenAI

GPT-4.1

1338