Vision Arena: LMArena Vision Arena
Crowdsourced human-preference leaderboard for vision-language (image + text input, text output) models run by LMArena. Voters compare two anonymous responses to the same image-and-text prompt and pick the better one; ratings use a Bradley-Terry model with 95% confidence intervals and style control (the default view since May 16, 2025). Higher is better.
Configurations
73
tracked on this benchmark
Score band
1309.0 – 993.0
best → lowest tracked
Snapshot trend
—
need ≥2 snapshots
Leaderboard
Tracked models ranked by Arena Score (higher is better).
Configuration: Claude Fable 5 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Fable 5 (High)': rank 1, Arena Score 1308.7 (95% CI 1301.7-1315.8), 14714 votes. Retrieved 2026-10-26.
Configuration: Qwen3.8-Max
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.8-Max': rank 2, Arena Score 1300.0 (95% CI 1292.5-1307.5), 9957 votes. Retrieved 2026-10-26.
Configuration: Claude Opus 4.6 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Opus 4.6 (High)': rank 3, Arena Score 1299.1 (95% CI 1292.6-1305.5), 22286 votes. Retrieved 2026-10-26.
Configuration: Claude Opus 4.7
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Opus 4.7': rank 4, Arena Score 1298.7 (95% CI 1292.1-1305.3), 23139 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.7 Flash (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.7 Flash (High)': rank 6, Arena Score 1295.2 (95% CI 1285.9-1304.4), 5369 votes. Retrieved 2026-10-26.
Configuration: Muse Spark
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Muse Spark': rank 7, Arena Score 1294.1 (95% CI 1285.0-1303.2), 5865 votes. Retrieved 2026-10-26.
Configuration: GPT-6.1 Sol (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-6.1 Sol (Max)': rank 9, Arena Score 1293.5 (95% CI 1280.6-1306.5), 2330 votes. Retrieved 2026-10-26.
Configuration: Claude Opus 5.5 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Opus 5.5 (High)': rank 10, Arena Score 1293.0 (95% CI 1281.2-1304.7), 3035 votes. Retrieved 2026-10-26.
Configuration: Muse Spark 1.2 (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Muse Spark 1.2 (xHigh)': rank 11, Arena Score 1292.7 (95% CI 1278.3-1307.0), 1948 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.8 Flash (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.8 Flash (High)': rank 12, Arena Score 1291.0 (95% CI 1281.2-1300.8), 4645 votes. Retrieved 2026-10-26.
Configuration: Muse Spark 1.3 (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Muse Spark 1.3 (Max)': rank 13, Arena Score 1289.6 (95% CI 1280.4-1298.8), 5392 votes. Retrieved 2026-10-26.
Configuration: Gemini 3 Pro
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3 Pro': rank 14, Arena Score 1289.3 (95% CI 1281.9-1296.7), 13623 votes. Retrieved 2026-10-26.
Configuration: Claude Opus 5 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Opus 5 (High)': rank 15, Arena Score 1287.9 (95% CI 1281.2-1294.5), 17743 votes. Retrieved 2026-10-26.
Configuration: Claude Fable 5.1 (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Fable 5.1 (Max)': rank 17, Arena Score 1287.6 (95% CI 1278.3-1296.9), 5026 votes. Retrieved 2026-10-26.
Configuration: GPT-5.6 Sol (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.6 Sol (xHigh)': rank 18, Arena Score 1286.2 (95% CI 1278.8-1293.5), 11439 votes. Retrieved 2026-10-26.
Configuration: GPT-5.5
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.5': rank 19, Arena Score 1286.0 (95% CI 1280.3-1291.8), 28781 votes. Retrieved 2026-10-26.
Configuration: Claude Opus 4.8 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Opus 4.8 (High)': rank 20, Arena Score 1285.7 (95% CI 1279.3-1292.2), 18303 votes. Retrieved 2026-10-26.
Configuration: GPT-6 Astra (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-6 Astra (Max)': rank 21, Arena Score 1284.8 (95% CI 1274.5-1295.1), 4012 votes. Retrieved 2026-10-26.
Configuration: GPT-5.4 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.4 (High)': rank 22, Arena Score 1284.1 (95% CI 1278.4-1289.9), 30747 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.5 Flash (Medium)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.5 Flash (Medium)': rank 23, Arena Score 1282.4 (95% CI 1275.6-1289.2), 14061 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.6 Flash (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.6 Flash (High)': rank 26, Arena Score 1282.0 (95% CI 1274.1-1289.9), 8979 votes. Retrieved 2026-10-26.
Configuration: Muse Spark 1.1
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Muse Spark 1.1': rank 27, Arena Score 1280.0 (95% CI 1273.0-1287.0), 12804 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.1 Pro Preview
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.1 Pro Preview': rank 28, Arena Score 1279.5 (95% CI 1274.4-1284.7), 47114 votes. Retrieved 2026-10-26.
Configuration: Grok 4.5
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Grok 4.5': rank 29, Arena Score 1279.0 (95% CI 1271.9-1286.1), 12599 votes. Retrieved 2026-10-26.
Configuration: GPT-5.5 Instant
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.5 Instant': rank 33, Arena Score 1276.1 (95% CI 1267.3-1284.8), 7590 votes. Retrieved 2026-10-26.
Configuration: Claude Sonnet 4.6
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Sonnet 4.6': rank 34, Arena Score 1275.1 (95% CI 1269.0-1281.2), 27408 votes. Retrieved 2026-10-26.
Configuration: Gemini 3 Flash
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3 Flash': rank 35, Arena Score 1272.5 (95% CI 1267.3-1277.7), 39131 votes. Retrieved 2026-10-26.
Configuration: GLM-5.3-Flash
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GLM-5.3-Flash': rank 36, Arena Score 1271.2 (95% CI 1264.8-1277.6), 16254 votes. Retrieved 2026-10-26.
Configuration: GPT-5.6 Terra (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.6 Terra (xHigh)': rank 37, Arena Score 1268.0 (95% CI 1260.8-1275.3), 11842 votes. Retrieved 2026-10-26.
Configuration: Gemini 3.5 Flash-Lite
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 3.5 Flash-Lite': rank 38, Arena Score 1266.9 (95% CI 1259.1-1274.8), 9238 votes. Retrieved 2026-10-26.
Configuration: GPT-6 Sol (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-6 Sol (Max)': rank 39, Arena Score 1266.0 (95% CI 1254.7-1277.4), 3221 votes. Retrieved 2026-10-26.
Configuration: Claude Sonnet 5.5 (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Sonnet 5.5 (xHigh)': rank 40, Arena Score 1264.9 (95% CI 1252.4-1277.5), 2564 votes. Retrieved 2026-10-26.
Configuration: Kimi K2.6
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Kimi K2.6': rank 41, Arena Score 1264.6 (95% CI 1257.8-1271.4), 16368 votes. Retrieved 2026-10-26.
Configuration: Qwen3.7-Plus
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.7-Plus': rank 42, Arena Score 1263.3 (95% CI 1256.6-1270.0), 14212 votes. Retrieved 2026-10-26.
Configuration: Grok 4.6 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Grok 4.6 (High)': rank 43, Arena Score 1262.1 (95% CI 1253.8-1270.5), 7261 votes. Retrieved 2026-10-26.
Configuration: Claude Sonnet 5 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Claude Sonnet 5 (High)': rank 44, Arena Score 1261.9 (95% CI 1255.0-1268.7), 14034 votes. Retrieved 2026-10-26.
Configuration: Gemma 4 31B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemma 4 31B': rank 45, Arena Score 1261.8 (95% CI 1256.1-1267.6), 40219 votes. Retrieved 2026-10-26.
Configuration: GPT-5.6 Luna (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.6 Luna (xHigh)': rank 48, Arena Score 1256.6 (95% CI 1249.2-1264.0), 11055 votes. Retrieved 2026-10-26.
Configuration: Kimi K2.5 (Thinking)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Kimi K2.5 (Thinking)': rank 51, Arena Score 1252.4 (95% CI 1246.7-1258.0), 28249 votes. Retrieved 2026-10-26.
Configuration: Mistral Large 4
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Mistral Large 4': rank 52, Arena Score 1251.9 (95% CI 1232.6-1271.3), 1008 votes. Retrieved 2026-10-26.
Configuration: GPT-5.1 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.1 (High)': rank 53, Arena Score 1251.2 (95% CI 1243.4-1259.0), 9848 votes. Retrieved 2026-10-26.
Configuration: GPT-5.4 mini (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.4 mini (High)': rank 54, Arena Score 1250.4 (95% CI 1244.2-1256.7), 25734 votes. Retrieved 2026-10-26.
Configuration: Grok 4.7 (xHigh)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Grok 4.7 (xHigh)': rank 56, Arena Score 1249.8 (95% CI 1239.1-1260.4), 3736 votes. Retrieved 2026-10-26.
Configuration: Gemini 2.5 Pro
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 2.5 Pro': rank 57, Arena Score 1247.6 (95% CI 1243.1-1252.2), 88755 votes. Retrieved 2026-10-26.
Configuration: Step 5 Preview
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Step 5 Preview': rank 58, Arena Score 1247.2 (95% CI 1232.0-1262.3), 1649 votes. Retrieved 2026-10-26.
Configuration: Qwen3.5-397B-A17B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.5-397B-A17B': rank 59, Arena Score 1245.8 (95% CI 1240.5-1251.1), 32657 votes. Retrieved 2026-10-26.
Configuration: MiMo-V2.6-Pro
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'MiMo-V2.6-Pro': rank 60, Arena Score 1244.6 (95% CI 1229.8-1259.3), 1754 votes. Retrieved 2026-10-26.
Configuration: MiMo-V2.6-Flash
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'MiMo-V2.6-Flash': rank 61, Arena Score 1244.5 (95% CI 1233.6-1255.4), 3618 votes. Retrieved 2026-10-26.
Configuration: GPT-5.2 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.2 (High)': rank 62, Arena Score 1242.8 (95% CI 1236.7-1248.9), 19022 votes. Retrieved 2026-10-26.
Configuration: Qwen3.8-27B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.8-27B': rank 63, Arena Score 1242.4 (95% CI 1234.2-1250.6), 7614 votes. Retrieved 2026-10-26.
Configuration: Grok 4.3
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Grok 4.3': rank 64, Arena Score 1242.3 (95% CI 1236.4-1248.1), 27901 votes. Retrieved 2026-10-26.
Configuration: Gemma 4 26B A4B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemma 4 26B A4B': rank 65, Arena Score 1241.7 (95% CI 1235.3-1248.2), 23922 votes. Retrieved 2026-10-26.
Configuration: GPT-6 Luna (Max)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-6 Luna (Max)': rank 66, Arena Score 1240.4 (95% CI 1229.5-1251.3), 3662 votes. Retrieved 2026-10-26.
Configuration: MiMo-V2.5
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'MiMo-V2.5': rank 71, Arena Score 1233.8 (95% CI 1227.9-1239.6), 29381 votes. Retrieved 2026-10-26.
Configuration: MiniMax-M3
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'MiniMax-M3': rank 72, Arena Score 1233.3 (95% CI 1226.9-1239.7), 19844 votes. Retrieved 2026-10-26.
Configuration: GLM-5V-Turbo
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GLM-5V-Turbo': rank 73, Arena Score 1230.0 (95% CI 1224.6-1235.5), 42757 votes. Retrieved 2026-10-26.
Configuration: Qwen3.5-122B-A10B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.5-122B-A10B': rank 76, Arena Score 1226.2 (95% CI 1219.4-1232.9), 13236 votes. Retrieved 2026-10-26.
Configuration: GPT-5 Chat
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5 Chat': rank 77, Arena Score 1225.8 (95% CI 1218.8-1232.9), 39894 votes. Retrieved 2026-10-26.
Configuration: Qwen3.5-27B
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3.5-27B': rank 80, Arena Score 1217.2 (95% CI 1211.6-1222.7), 30272 votes. Retrieved 2026-10-26.
Configuration: MiMo-V2-Omni
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'MiMo-V2-Omni': rank 81, Arena Score 1215.9 (95% CI 1208.5-1223.3), 10650 votes. Retrieved 2026-10-26.
Configuration: Gemini 2.5 Flash
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Gemini 2.5 Flash': rank 83, Arena Score 1215.5 (95% CI 1210.9-1220.0), 60087 votes. Retrieved 2026-10-26.
Configuration: Qwen3-VL-235B-A22B-Instruct
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Qwen3-VL-235B-A22B-Instruct': rank 84, Arena Score 1215.1 (95% CI 1208.1-1222.1), 12724 votes. Retrieved 2026-10-26.
Configuration: GPT-5 (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5 (High)': rank 86, Arena Score 1210.7 (95% CI 1203.7-1217.8), 34445 votes. Retrieved 2026-10-26.
Configuration: Inkling-Small
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Inkling-Small': rank 87, Arena Score 1208.8 (95% CI 1200.6-1216.9), 8074 votes. Retrieved 2026-10-26.
Configuration: Mistral Medium 3.5
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Mistral Medium 3.5': rank 93, Arena Score 1199.6 (95% CI 1190.0-1209.2), 5375 votes. Retrieved 2026-10-26.
Configuration: GPT-5.4 nano (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5.4 nano (High)': rank 95, Arena Score 1198.8 (95% CI 1192.5-1205.2), 25624 votes. Retrieved 2026-10-26.
Configuration: Grok 4.1 Fast (Reasoning)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Grok 4.1 Fast (Reasoning)': rank 96, Arena Score 1196.3 (95% CI 1189.4-1203.3), 13368 votes. Retrieved 2026-10-26.
Configuration: GPT-5 mini (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5 mini (High)': rank 104, Arena Score 1184.3 (95% CI 1176.2-1192.4), 28943 votes. Retrieved 2026-10-26.
Configuration: GLM-4.6V
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GLM-4.6V': rank 111, Arena Score 1161.4 (95% CI 1147.6-1175.2), 2455 votes. Retrieved 2026-10-26.
Configuration: GLM-4.5V
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GLM-4.5V': rank 118, Arena Score 1153.1 (95% CI 1141.3-1164.8), 3243 votes. Retrieved 2026-10-26.
Configuration: GPT-5 nano (High)
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'GPT-5 nano (High)': rank 120, Arena Score 1147.4 (95% CI 1136.1-1158.8), 3928 votes. Retrieved 2026-10-26.
Configuration: Llama 3.2 90B Vision Instruct
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Llama 3.2 90B Vision Instruct': rank 148, Arena Score 1032.2 (95% CI 1023.0-1041.5), 8682 votes. Retrieved 2026-10-26.
Configuration: Llama 3.2 11B Vision Instruct
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'Llama 3.2 11B Vision Instruct': rank 159, Arena Score 992.7 (95% CI 981.4-1004.1), 4817 votes. Retrieved 2026-10-26.
Attribution. Source: LMArena (lmarena.ai), CC BY 4.0
How to read this benchmark
This benchmark scores models where higher is better. Use scores for directional filtering and shortlisting, not universal quality ranking; then validate pricing, context window, provider availability, and fit for your workload.
Trust this score when
- The row names the exact model configuration, benchmark version, evaluator, harness, and observation date.
- Cost, token, and step averages are evaluated together with the score for the same configuration.
- You validate the configuration in your own Cursor workload before making a model choice.
Be cautious when
- Rows are collapsed into one score per base model or compared across benchmark versions.
- Small score differences are treated as decisive despite Cursor's variance caveat.
- A qualified row, including Grok 4.5's disclosed training-data advantage, is used as neutral winner evidence.
Related benchmarks
Last reviewed: Oct 26, 2026