← Arena leaderboards

Vision · Creative Writing Vision

Arena · 50 models

RankModelScore
1Gemini 3 Pro1,341.5684
2Claude Fable 5 (High)1,337.6994
3Claude Opus 4.6 (High)1,331.3013
4Claude Fable 5.1 (Max)1,329.7188
5Muse Spark1,326.9924
6Claude Opus 4.7 (High)1,325.5696
7Gemini 3.1 Pro Preview1,324.1944
8Claude Opus 4.71,320.9147
9Claude Opus 5.5 (High)1,320.4063
10Claude Opus 4.61,319.4988
11Gemini 3 Flash1,314.2168
12Claude Opus 5 (High)1,312.9301
13Qwen3.8-Max1,311.6866
14Gemini 3.7 Flash (High)1,311.5314
15GPT-5.5 (High)1,301.3554
16Gemini 3.5 Flash (High)1,301.2991
17Gemini 3.5 Flash (Medium)1,301.1734
18Claude Opus 4.8 (High)1,300.8985
19GPT-6.1 Sol (Max)1,300.6644
20Gemini 3.8 Flash (High)1,299.2678
21Muse Spark 1.3 (Max)1,295.7026
22Gemini 3 Flash (Minimal)1,293.3037
23Grok 4.51,292.691
24Claude Opus 4.81,291.7438
25GPT-5.51,287.1776
26Muse Spark 1.3 (xHigh)1,285.9383
27GPT-6 Astra (Max)1,285.5228
28Gemma 4 31B1,284.7285
29GPT-5.6 Sol (xHigh)1,283.3911
30GPT-5.4 (High)1,283.1181
31Grok 4.20 Multi-Agent Beta (0309)1,280.8625
32Gemini 3.6 Flash (High)1,278.3802
33Grok 4.20 Beta (Reasoning, 0309)1,275.6194
34DeepSeek-V4.1-Flash-20260910 (Max)1,274.612
35Kimi K2.61,273.2003
36Claude Sonnet 4.61,272.3156
37GPT-5.41,272.0583
38Gemini 2.5 Flash Preview (09-2025)1,270.1157
39GLM-5.3-Flash1,269.5566
40Kimi K2.5 (Thinking)1,268.7803
41GPT-5.11,266.3542
42Gemini 2.5 Pro1,264.7158
43Grok 4.6 (High)1,262.1281
44Qwen3.5-397B-A17B1,260.3034
45Gemini 3.5 Flash-Lite1,259.7209
46GPT-5.5 Instant1,259.4458
47Dola Seed 2.0 Pro1,258.3597
48Qwen3.7-Plus1,257.7104
49GPT-5.6 Luna (xHigh)1,256.7731
50Gemma 4 26B A4B1,255.7526