MODEL INTELLIGENCE
ModelRank
Compare models within one task and one source. Human preference, benchmark ability, and usage measure different things; their scores are not combined.
Methodology and data notes
Vision arena with style control.
Metric: Style-controlled Arena rating, higher is better
Ranking source
| Rank | Model | Organization | Style-controlled Arena rating ↑ | 95% CI | Access | Evidence |
|---|---|---|---|---|---|---|
| 1 |  Claude Fable 5 (High) Style-controlled Arena rating ↑ | anthropic | 1,308.7 | 1301.7–1315.8 | Proprietary | 14,714 votes Source |
| 2 | Qwen3.8-Max Style-controlled Arena rating ↑ | alibaba | 1,300.0 | 1292.5–1307.5 | Proprietary | 9,957 votes Source |
| 3 | Claude Opus 4.6 (High) Style-controlled Arena rating ↑ | anthropic | 1,299.1 | 1292.6–1305.5 | Proprietary | 22,286 votes Source |
| 4 | Claude Opus 4.7 Style-controlled Arena rating ↑ | anthropic | 1,298.7 | 1292.1–1305.3 | Proprietary | 23,139 votes Source |
| 5 | Claude Opus 4.7 (High) Style-controlled Arena rating ↑ | anthropic | 1,297.7 | 1291.2–1304.3 | Proprietary | 22,731 votes Source |
| 6 |  Gemini 3.7 Flash (High) Style-controlled Arena rating ↑ | 1,295.2 | 1285.9–1304.4 | Proprietary | 5,369 votes Source | |
| 7 |  Muse Spark Style-controlled Arena rating ↑ | meta | 1,294.1 | 1285.0–1303.2 | Proprietary | 5,865 votes Source |
| 8 | Claude Opus 4.6 Style-controlled Arena rating ↑ | anthropic | 1,293.7 | 1287.4–1300.0 | Proprietary | 26,841 votes Source |
| 9 |  GPT-6.1 Sol (Max) Style-controlled Arena rating ↑ | openai | 1,293.5 | 1280.6–1306.5 | Proprietary | 2,330 votes Source |
| 10 | Claude Opus 5.5 (High) Style-controlled Arena rating ↑ | anthropic | 1,293.0 | 1281.2–1304.7 | Proprietary | 3,035 votes Source |
| 11 | Muse Spark 1.2 (xHigh) Style-controlled Arena rating ↑ | meta | 1,292.7 | 1278.3–1307.0 | Proprietary | 1,948 votes Source |
| 12 | Gemini 3.8 Flash (High) Style-controlled Arena rating ↑ | 1,291.0 | 1281.2–1300.8 | Proprietary | 4,645 votes Source | |
| 13 | Muse Spark 1.3 (Max) Style-controlled Arena rating ↑ | meta | 1,289.6 | 1280.4–1298.8 | Proprietary | 5,392 votes Source |
| 14 | Gemini 3 Pro Style-controlled Arena rating ↑ | 1,289.3 | 1281.9–1296.7 | Proprietary | 13,623 votes Source | |
| 15 | Claude Opus 5 (High) Style-controlled Arena rating ↑ | anthropic | 1,287.9 | 1281.2–1294.5 | Proprietary | 17,743 votes Source |
| 16 | Muse Spark 1.3 (xHigh) Style-controlled Arena rating ↑ | meta | 1,287.6 | 1278.1–1297.2 | Proprietary | 4,897 votes Source |
| 17 | Claude Fable 5.1 (Max) Style-controlled Arena rating ↑ | anthropic | 1,287.6 | 1278.3–1296.9 | Proprietary | 5,026 votes Source |
| 18 | GPT-5.6 Sol (xHigh) Style-controlled Arena rating ↑ | openai | 1,286.2 | 1278.8–1293.5 | Proprietary | 11,439 votes Source |
| 19 | GPT-5.5 Style-controlled Arena rating ↑ | openai | 1,286.0 | 1280.3–1291.8 | Proprietary | 28,781 votes Source |
| 20 | Claude Opus 4.8 (High) Style-controlled Arena rating ↑ | anthropic | 1,285.7 | 1279.3–1292.2 | Proprietary | 18,303 votes Source |
| 21 | GPT-6 Astra (Max) Style-controlled Arena rating ↑ | openai | 1,284.8 | 1274.5–1295.1 | Proprietary | 4,012 votes Source |
| 22 | GPT-5.4 (High) Style-controlled Arena rating ↑ | openai | 1,284.1 | 1278.4–1289.9 | Proprietary | 30,747 votes Source |
| 23 | Gemini 3.5 Flash (Medium) Style-controlled Arena rating ↑ | 1,282.4 | 1275.6–1289.2 | Proprietary | 14,061 votes Source | |
| 24 | Gemini 3.5 Flash (High) Style-controlled Arena rating ↑ | 1,282.2 | 1275.4–1289.1 | Proprietary | 13,862 votes Source | |
| 25 | GPT-5.5 (High) Style-controlled Arena rating ↑ | openai | 1,282.2 | 1276.3–1288.0 | Proprietary | 27,067 votes Source |
| 26 | Gemini 3.6 Flash (High) Style-controlled Arena rating ↑ | 1,282.0 | 1274.1–1289.9 | Proprietary | 8,979 votes Source | |
| 27 | Muse Spark 1.1 Style-controlled Arena rating ↑ | meta | 1,280.0 | 1273.0–1287.0 | Proprietary | 12,804 votes Source |
| 28 | Gemini 3.1 Pro Preview Style-controlled Arena rating ↑ | 1,279.5 | 1274.4–1284.7 | Proprietary | 47,114 votes Source | |
| 29 | Grok 4.5 Style-controlled Arena rating ↑ | xai | 1,279.0 | 1271.9–1286.1 | Proprietary | 12,599 votes Source |
| 30 | Claude Opus 4.8 Style-controlled Arena rating ↑ | anthropic | 1,278.9 | 1272.4–1285.4 | Proprietary | 18,800 votes Source |
| 31 | GPT-5.4 Style-controlled Arena rating ↑ | openai | 1,278.6 | 1272.1–1285.1 | Proprietary | 22,804 votes Source |
| 32 | GPT-5.2 Chat (20260210) Style-controlled Arena rating ↑ | openai | 1,277.7 | 1271.2–1284.2 | Proprietary | 16,465 votes Source |
| 33 | GPT-5.5 Instant Style-controlled Arena rating ↑ | openai | 1,276.1 | 1267.3–1284.8 | Proprietary | 7,590 votes Source |
| 34 | Claude Sonnet 4.6 Style-controlled Arena rating ↑ | anthropic | 1,275.1 | 1269.0–1281.2 | Proprietary | 27,408 votes Source |
| 35 | Gemini 3 Flash Style-controlled Arena rating ↑ | 1,272.5 | 1267.3–1277.7 | Proprietary | 39,131 votes Source | |
| 36 | GLM-5.3-Flash Style-controlled Arena rating ↑ | zai | 1,271.2 | 1264.8–1277.6 | MIT | 16,254 votes Source |
| 37 | GPT-5.6 Terra (xHigh) Style-controlled Arena rating ↑ | openai | 1,268.0 | 1260.8–1275.3 | Proprietary | 11,842 votes Source |
| 38 | Gemini 3.5 Flash-Lite Style-controlled Arena rating ↑ | 1,266.9 | 1259.1–1274.8 | Proprietary | 9,238 votes Source | |
| 39 | GPT-6 Sol (Max) Style-controlled Arena rating ↑ | openai | 1,266.0 | 1254.7–1277.4 | Proprietary | 3,221 votes Source |
| 40 | Claude Sonnet 5.5 (xHigh) Style-controlled Arena rating ↑ | anthropic | 1,264.9 | 1252.4–1277.5 | Proprietary | 2,564 votes Source |
| 41 | Kimi K2.6 Style-controlled Arena rating ↑ | moonshot | 1,264.6 | 1257.8–1271.4 | Modified MIT | 16,368 votes Source |
| 42 | Qwen3.7-Plus Style-controlled Arena rating ↑ | alibaba | 1,263.3 | 1256.6–1270.0 | Proprietary | 14,212 votes Source |
| 43 | Grok 4.6 (High) Style-controlled Arena rating ↑ | xai | 1,262.1 | 1253.8–1270.5 | Proprietary | 7,261 votes Source |
| 44 | Claude Sonnet 5 (High) Style-controlled Arena rating ↑ | anthropic | 1,261.9 | 1255.0–1268.7 | Proprietary | 14,034 votes Source |
| 45 | Gemma 4 31B Style-controlled Arena rating ↑ | 1,261.8 | 1256.1–1267.6 | Apache 2.0 | 40,219 votes Source | |
| 46 | DeepSeek-V4.1-Flash-20260910 (Max) Style-controlled Arena rating ↑ | deepseek | 1,261.4 | 1251.1–1271.6 | MIT | 4,100 votes Source |
| 47 | Gemini 3 Flash (Minimal) Style-controlled Arena rating ↑ | 1,260.0 | 1254.6–1265.3 | Proprietary | 37,173 votes Source | |
| 48 | GPT-5.6 Luna (xHigh) Style-controlled Arena rating ↑ | openai | 1,256.6 | 1249.2–1264.0 | Proprietary | 11,055 votes Source |
| 49 |  Dola Seed 2.0 Pro Style-controlled Arena rating ↑ | bytedance | 1,256.5 | 1249.2–1263.7 | Proprietary | 10,972 votes Source |
| 50 | Grok 4.20 Beta (Reasoning, 0309) Style-controlled Arena rating ↑ | xai | 1,255.2 | 1249.1–1261.3 | Proprietary | 26,796 votes Source |