llm.ing

LMArena Vision

crowdrun by LMArenaunit: eloCC-BY-4.057 modelssource ↗methodology ↗

Caveat: Image-understanding battles.

TextWebDevVision
#ModelScoreObserved
1Claude Fable 5
1326
±8· 11.3K votes
Sep 13, 2026
2Claude Opus 5high-effort
1321
±8· 11.6K votes
Sep 13, 2026
3Claude Opus 4.7
1318
±7· 21.5K votes
Sep 13, 2026
4Claude Opus 4.6thinking
1316
±7· 20.3K votes
Aug 6, 2026
5Qwen3.8 Max
1315
±8· 8.7K votes
Sep 13, 2026
6Gemini 3.6 Flash
1311
±18· 1.2K votes
Aug 6, 2026
7Gemini 3.5 Flashhigh-effort
1310
±8· 9.6K votes
Sep 13, 2026
8GPT-5.4high-effort
1302
±6· 25.4K votes
Sep 13, 2026
9GLM-5.3-Flash
1299
±12· 3.1K votes
Sep 13, 2026
10GPT-5.5
1298
±6· 23.4K votes
Sep 13, 2026
11Gemini 3.1 Pro Preview
1296
±5· 40.7K votes
Sep 13, 2026
12Muse Spark 1.1
1294
±8· 8.4K votes
Sep 13, 2026
13Claude Opus 4.8thinking
1293
±8· 11.1K votes
Aug 6, 2026
14Grok 4.5
1289
±8· 8.4K votes
Sep 13, 2026
15Claude Sonnet 4.6
1283
±6· 25.6K votes
Sep 13, 2026
16Kimi K2.6
1280
±7· 15.3K votes
Sep 13, 2026
17GPT-5.6 Solxhigh-effort
1280
±8· 7.7K votes
Sep 13, 2026
18Qwen3.7 Plus
1278
±8· 10.7K votes
Sep 13, 2026
19Claude Sonnet 5high-effort
1277
±8· 10.3K votes
Sep 13, 2026
20GPT-5.6 Terraxhigh-effort
1270
±8· 7.7K votes
Sep 13, 2026
21Gemini 3.5 Flash Lite
1268
±10· 5.3K votes
Sep 13, 2026
22GPT-5.2 Chat
1268
±7· 15.4K votes
Sep 13, 2026
23Kimi K2.5thinking
1267
±6· 26.6K votes
Sep 13, 2026
24Grok 4.6high-effort
1267
±11· 3.6K votes
Sep 13, 2026
25Qwen3.5 397B-A17B
1265
±6· 27.6K votes
Sep 13, 2026
26GLM-5V-Turbo
1262
±6· 37.4K votes
Sep 13, 2026
27Gemini 2.5 Pro
1262
±5· 87.6K votes
Sep 13, 2026
28GPT-5.6 Lunaxhigh-effort
1259
±8· 7.8K votes
Sep 13, 2026
29MiniMax-M3
1254
±7· 15.2K votes
Sep 13, 2026
30GPT-5.1high-effort
1248
±8· 9.3K votes
Sep 13, 2026
31GPT-5.2high-effort
1247
±6· 17.9K votes
Sep 13, 2026
32Qwen3.5 122B-A10B
1245
±7· 12.5K votes
Sep 13, 2026
33GPT-5.4 minihigh-effort
1245
±6· 24K votes
Sep 13, 2026
34Qwen3.5 27B
1241
±6· 25.4K votes
Sep 13, 2026
35Gemini 3.1 Flash Lite Preview
1241
±6· 26K votes
Sep 13, 2026
36Gemini 2.5 Flash
1236
±5· 58.9K votes
Sep 13, 2026
37Grok 4.3
1229
±6· 22.7K votes
Sep 13, 2026
38o3
1214
±7· 45.6K votes
Sep 13, 2026
39Qwen-VL Max
1212
±12· 3.2K votes
Sep 13, 2026
40GPT-4.1
1210
±7· 41K votes
Sep 13, 2026
41GPT-5high-effort
1209
±7· 34.4K votes
Sep 13, 2026
42Qwen3-VL 235B-A22Bthinking
1206
±12· 2.4K votes
Sep 13, 2026
43GPT-5 Minihigh-effort
1202
±8· 29K votes
Sep 13, 2026
44GPT-5.4 nanohigh-effort
1198
±7· 23.9K votes
Sep 13, 2026
45o4-mini
1193
±7· 41.3K votes
Sep 13, 2026
46GPT-4.1 mini
1181
±7· 40.3K votes
Sep 13, 2026
47o1
1168
±11· 3.7K votes
Sep 13, 2026
48GLM-4.6V
1163
±14· 2.4K votes
Sep 13, 2026
49GPT-5 Nanohigh-effort
1159
±11· 3.9K votes
Sep 13, 2026
50GLM-4.5V
1155
±12· 3.3K votes
Sep 13, 2026
51GPT-4o
1137
±9· 23.3K votes
Sep 13, 2026
52Qwen2.5-VL 72B Instruct
1108
±11· 3.8K votes
Aug 6, 2026
53GPT-4 Turbo
1090
±12· 13.4K votes
Sep 13, 2026
54GPT-4o mini
1066
±8· 17.3K votes
Sep 13, 2026
55GPT-4.1 nano
1063
±18· 1.2K votes
Sep 13, 2026
56Pixtral 12B
1009
±10· 7.5K votes
Sep 13, 2026
57Aya Vision 32B
996
±22· 847 votes
Sep 13, 2026

Dot = elo on a 955–1352 scale (position, not length); whisker = rating confidence interval. Scores link to model pages; every number keeps its source on the model page.