LMArena snapshot · published July 16, 2026
Anthropic leads this LMArena snapshot.
This leaderboard ranks every resolved model in one published LMArena rating for text models, using LMArena's official ranks. A second table summarizes the leading model from each of seven organizations. Confidence intervals and battle counts appear alongside each score.
Model rankings · LMArena overall
Every resolved model, ranked.
| Rank | LMArena rating | Model | Organization | Reported interval | Gap to leader | Battles |
|---|---|---|---|---|---|---|
| 01 | 1506.8 | Claude Fable 5 | Anthropic | 1499.7 to 1513.8 | Leader | 9,798 |
| 02 | 1504.2 | Claude Opus 4.6 Thinking | Anthropic | 1500.4 to 1507.9 | -2.6 | 62,355 |
| 03 | 1501.9 | Claude Opus 4.7 Thinking | Anthropic | 1497.7 to 1506.1 | -4.9 | 49,798 |
| 04 | 1497.9 | Claude Opus 4.6 | Anthropic | 1494.3 to 1501.6 | -8.9 | 66,131 |
| 05 | 1495.1 | Muse Spark 1.1 | Meta | 1487.6 to 1502.6 | -11.7 | 7,074 |
| 06 | 1493.7 | Claude Opus 4.7 | Anthropic | 1489.5 to 1497.9 | -13.1 | 50,890 |
| 07 | 1487.4 | Muse Spark | Meta | 1481.5 to 1493.2 | -19.4 | 13,562 |
| 09 | 1486.4 | GPT-5.6 Sol X-High | OpenAI | 1478.0 to 1494.7 | -20.4 | 5,467 |
| 10 | 1485.8 | Gemini 3 Pro | 1481.9 to 1489.6 | -21.0 | 41,279 | |
| 11 | 1485.6 | Gemini 3.1 Pro Preview | 1482.1 to 1489.1 | -21.2 | 83,386 | |
| 12 | 1484.6 | Claude Opus 4.8 Thinking | Anthropic | 1479.5 to 1489.8 | -22.2 | 30,024 |
| 13 | 1481.0 | GPT-5.5 High | OpenAI | 1476.6 to 1485.4 | -25.8 | 44,927 |
| 14 | 1477.9 | GPT-5.4 High | OpenAI | 1473.9 to 1481.8 | -28.9 | 58,116 |
| 15 | 1476.3 | Gemini 3.5 Flash High | 1469.8 to 1482.9 | -30.5 | 10,094 | |
| 16 | 1475.7 | GPT-5.2 Chat Latest (2026-02-10) | OpenAI | 1471.6 to 1479.9 | -31.1 | 34,416 |
| 17 | 1475.7 | GPT-5.5 | OpenAI | 1471.3 to 1480.1 | -31.1 | 46,328 |
| 18 | 1475.3 | Qwen3.7 Max Preview | Alibaba | 1465.3 to 1485.4 | -31.5 | 3,715 |
| 19 | 1474.8 | Gemini 3.5 Flash Medium | 1468.7 to 1481.0 | -32.0 | 13,158 | |
| 20 | 1474.4 | Claude Opus 4.8 | Anthropic | 1469.3 to 1479.6 | -32.4 | 30,670 |
| 21 | 1474.2 | Grok 4.20 Beta 1 | xAI | 1469.5 to 1478.8 | -32.6 | 26,827 |
| 22 | 1473.3 | Grok 4.20 Beta 0309 Reasoning | xAI | 1469.4 to 1477.1 | -33.5 | 59,422 |
| 23 | 1473.2 | GPT-5.5 Instant | OpenAI | 1468.1 to 1478.3 | -33.6 | 26,015 |
| 24 | 1473.0 | Gemini 3 Flash | 1468.6 to 1477.4 | -33.8 | 30,695 | |
| 25 | 1473.0 | Claude Opus 4.5 Thinking 32K (2025-11-01) | Anthropic | 1469.1 to 1476.9 | -33.8 | 37,049 |
| 26 | 1472.5 | Claude Sonnet 4.6 | Anthropic | 1468.6 to 1476.3 | -34.3 | 56,335 |
Show the remaining 299 resolved rows
| Rank | LMArena rating | Model | Organization | Reported interval | Gap to leader | Battles |
|---|---|---|---|---|---|---|
| 27 | 1470.4 | Grok 4.20 Multi Agent Beta 0309 | xAI | 1466.5 to 1474.3 | -36.4 | 58,221 |
| 28 | 1469.9 | GLM-5.1 | Z.ai | 1465.3 to 1474.5 | -36.9 | 29,919 |
| 29 | 1469.5 | GLM 5.2 (Max) | Z.ai | 1463.6 to 1475.5 | -37.3 | 17,103 |
| 30 | 1469.2 | Claude Opus 4.5 (2025-11-01) | Anthropic | 1466.0 to 1472.4 | -37.6 | 70,973 |
| 31 | 1467.9 | Ernie 5.1 | Baidu | 1463.2 to 1472.5 | -38.9 | 37,445 |
| 32 | 1466.7 | GPT 5.4 | OpenAI | 1462.8 to 1470.6 | -40.1 | 61,064 |
| 33 | 1465.9 | Mimo V2.5 Pro | Xiaomi | 1461.4 to 1470.4 | -40.9 | 41,409 |
| 34 | 1465.9 | Grok 4.1 Thinking | xAI | 1462.7 to 1469.1 | -40.9 | 65,461 |
| 35 | 1465.7 | Grok 4.5 | xAI | 1458.1 to 1473.3 | -41.1 | 6,942 |
| 36 | 1464.8 | Qwen3.5 Max Preview | Alibaba | 1459.8 to 1469.8 | -42.0 | 21,479 |
| 37 | 1461.4 | Claude Sonnet 5 High | Anthropic | 1455.1 to 1467.8 | -45.4 | 12,645 |
| 38 | 1461.2 | Kimi K2.6 | Moonshot AI | 1456.6 to 1465.7 | -45.6 | 37,711 |
| 39 | 1460.8 | Qwen3.7 Plus | Alibaba | 1455.1 to 1466.4 | -46.0 | 21,513 |
| 40 | 1460.3 | Qwen3.6 Max Preview | Alibaba | 1451.9 to 1468.7 | -46.5 | 5,191 |
| 41 | 1459.5 | Grok 4.1 | xAI | 1456.2 to 1462.7 | -47.3 | 67,587 |
| 42 | 1458.8 | Gemini 3 Flash (thinking Minimal) | 1455.6 to 1461.9 | -48.0 | 83,526 | |
| 43 | 1457.1 | DeepSeek V4 Pro | DeepSeek | 1452.7 to 1461.5 | -49.7 | 44,418 |
| 44 | 1456.7 | Glm 5 | Z.ai | 1452.3 to 1461.0 | -50.1 | 27,789 |
| 45 | 1456.0 | DeepSeek V4 Pro Thinking | DeepSeek | 1451.5 to 1460.5 | -50.8 | 42,294 |
| 46 | 1456.0 | Dola Seed 2.0 Pro | ByteDance | 1452.3 to 1459.6 | -50.8 | 67,083 |
| 47 | 1455.5 | Claude Sonnet 4 5 20250929 Thinking 32k | Anthropic | 1452.7 to 1458.3 | -51.3 | 82,325 |
| 48 | 1455.1 | Claude Sonnet 4 5 20250929 | Anthropic | 1452.2 to 1458.0 | -51.7 | 80,727 |
| 49 | 1454.7 | GPT 5.1 High | OpenAI | 1450.9 to 1458.4 | -52.1 | 40,779 |
| 50 | 1450.9 | Gemma 4 31b | 1443.3 to 1458.5 | -55.9 | 5,879 | |
| 51 | 1449.3 | GPT 5.4 Mini High | OpenAI | 1445.4 to 1453.3 | -57.5 | 56,941 |
| 52 | 1449.2 | Kimi K2.5 Thinking | Moonshot AI | 1445.6 to 1452.7 | -57.6 | 61,978 |
| 53 | 1449.1 | Claude Opus 4 1 20250805 Thinking 16k | Anthropic | 1445.7 to 1452.6 | -57.7 | 49,757 |
| 54 | 1448.8 | GPT 5.3 Chat Latest | OpenAI | 1444.5 to 1453.2 | -58.0 | 32,982 |
| 55 | 1448.8 | Ernie 5.0 Preview 1203 | Baidu | 1442.3 to 1455.3 | -58.0 | 9,734 |
| 56 | 1448.3 | Mimo V2 Pro | Xiaomi | 1443.6 to 1453.1 | -58.5 | 24,478 |
| 57 | 1447.0 | Claude Opus 4 1 20250805 | Anthropic | 1444.1 to 1450.0 | -59.8 | 77,243 |
| 58 | 1446.8 | Ernie 5.0 0110 | Baidu | 1442.9 to 1450.8 | -60.0 | 35,225 |
| 60 | 1445.6 | Gemini 2.5 Pro | 1443.1 to 1448.1 | -61.2 | 124,385 | |
| 61 | 1445.0 | Minimax M3 | MiniMax | 1439.7 to 1450.4 | -61.8 | 27,238 |
| 62 | 1444.7 | GPT 4.5 Preview 2025 02 27 | OpenAI | 1439.0 to 1450.4 | -62.1 | 14,547 |
| 63 | 1443.5 | Qwen3.6 Plus | Alibaba | 1439.2 to 1447.8 | -63.3 | 43,163 |
| 64 | 1443.1 | Chatgpt 4o Latest 20250326 | OpenAI | 1440.3 to 1445.9 | -63.7 | 82,388 |
| 65 | 1442.7 | Grok 4.3 | xAI | 1438.3 to 1447.0 | -64.1 | 45,525 |
| 66 | 1442.2 | Qwen3.5 397b A17b | Alibaba | 1438.5 to 1446.0 | -64.6 | 57,410 |
| 67 | 1442.1 | Glm 4.7 | Z.ai | 1436.0 to 1448.2 | -64.7 | 12,094 |
| 68 | 1438.6 | GPT 5.1 | OpenAI | 1435.0 to 1442.3 | -68.2 | 43,398 |
| 69 | 1438.3 | DeepSeek V4 Flash Thinking | DeepSeek | 1433.9 to 1442.7 | -68.5 | 44,034 |
| 70 | 1438.2 | Gemma 4 26b A4b | 1430.6 to 1445.9 | -68.6 | 5,798 | |
| 71 | 1437.4 | GPT 5.2 High | OpenAI | 1433.7 to 1441.0 | -69.4 | 47,939 |
| 72 | 1436.5 | Glm 5v Turbo | Z.ai | 1428.8 to 1444.1 | -70.3 | 6,753 |
| 73 | 1436.1 | DeepSeek V4 Flash | DeepSeek | 1431.8 to 1440.5 | -70.7 | 44,220 |
| 74 | 1435.8 | Longcat Flash Chat 2602 Exp | Meituan | 1431.1 to 1440.4 | -71.0 | 28,080 |
| 75 | 1434.8 | Qwen3 Max Preview | Alibaba | 1430.3 to 1439.3 | -72.0 | 27,696 |
| 76 | 1434.5 | GPT 5.2 | OpenAI | 1431.2 to 1437.8 | -72.3 | 76,740 |
| 77 | 1433.8 | GPT 5 High | OpenAI | 1429.3 to 1438.3 | -73.0 | 31,889 |
| 78 | 1433.1 | MiMo V2.5 | Xiaomi | 1428.6 to 1437.5 | -73.7 | 42,281 |
| 79 | 1431.8 | Gemini 3.1 Flash Lite Preview | 1428.0 to 1435.5 | -75.0 | 60,792 | |
| 80 | 1431.6 | Kimi K2.5 Instant | Moonshot AI | 1425.1 to 1438.2 | -75.2 | 8,177 |
| 81 | 1431.0 | Grok 4 1 Fast Reasoning | xAI | 1427.7 to 1434.2 | -75.8 | 56,767 |
| 82 | 1430.9 | O3 2025 04 16 | OpenAI | 1427.3 to 1434.5 | -75.9 | 59,695 |
| 83 | 1430.1 | Mimo V2 Omni | Xiaomi | 1424.3 to 1435.8 | -76.7 | 19,464 |
| 84 | 1429.6 | Kimi K2 Thinking Turbo | Moonshot AI | 1426.5 to 1432.8 | -77.2 | 61,948 |
| 85 | 1427.6 | Mistral Medium 3.5 | Mistral AI | 1421.0 to 1434.2 | -79.2 | 11,011 |
| 86 | 1426.9 | Amazon Nova Experimental Chat 26 02 10 | Amazon | 1417.1 to 1436.8 | -79.9 | 3,419 |
| 87 | 1426.7 | GPT 5 Chat | OpenAI | 1422.4 to 1431.0 | -80.1 | 31,519 |
| 88 | 1425.2 | Glm 4.6 | Z.ai | 1421.3 to 1429.1 | -81.6 | 35,613 |
| 89 | 1424.9 | DeepSeek V3.2 | DeepSeek | 1421.3 to 1428.5 | -81.9 | 47,211 |
| 90 | 1424.7 | DeepSeek V3.2 Exp Thinking | DeepSeek | 1418.1 to 1431.2 | -82.1 | 9,067 |
| 91 | 1424.5 | Claude Opus 4 20250514 Thinking 16k | Anthropic | 1420.1 to 1428.8 | -82.3 | 36,860 |
| 92 | 1424.5 | Nvidia Nemotron 3 Ultra 550b A55b Nvfp4 | NVIDIA | 1417.4 to 1431.6 | -82.3 | 10,283 |
| 93 | 1424.1 | Qwen3 Max 2025 09 23 | Alibaba | 1417.6 to 1430.5 | -82.7 | 9,148 |
| 94 | 1423.1 | Qwen3 235b A22b Instruct 2507 | Alibaba | 1420.5 to 1425.7 | -83.7 | 97,091 |
| 95 | 1422.9 | DeepSeek V3.2 Thinking | DeepSeek | 1419.3 to 1426.6 | -83.9 | 41,026 |
| 96 | 1422.7 | DeepSeek V3.2 Exp | DeepSeek | 1416.3 to 1429.1 | -84.1 | 11,914 |
| 97 | 1422.1 | DeepSeek R1 0528 | DeepSeek | 1416.4 to 1427.7 | -84.7 | 18,452 |
| 98 | 1420.8 | Grok 4 Fast Chat | xAI | 1413.1 to 1428.4 | -86.0 | 6,807 |
| 99 | 1418.4 | Ernie 5.0 Preview 1022 | Baidu | 1409.6 to 1427.2 | -88.4 | 4,702 |
| 100 | 1417.9 | Kimi K2 0905 Preview | Moonshot AI | 1411.4 to 1424.4 | -88.9 | 11,771 |
| 101 | 1417.8 | Minimax M2.7 | MiniMax | 1413.6 to 1422.0 | -89.0 | 49,135 |
| 102 | 1417.5 | Kimi K2 0711 Preview | Moonshot AI | 1412.6 to 1422.4 | -89.3 | 27,605 |
| 103 | 1417.5 | DeepSeek V3.1 | DeepSeek | 1411.5 to 1423.5 | -89.3 | 14,941 |
| 104 | 1417.4 | DeepSeek V3.1 Terminus Thinking | DeepSeek | 1407.4 to 1427.4 | -89.4 | 3,456 |
| 105 | 1417.3 | Qwen3.5 122b A10b | Alibaba | 1412.9 to 1421.7 | -89.5 | 28,501 |
| 106 | 1417.0 | DeepSeek V3.1 Thinking | DeepSeek | 1410.4 to 1423.6 | -89.8 | 11,723 |
| 107 | 1415.4 | Amazon Nova Experimental Chat 26 01 10 | Amazon | 1405.5 to 1425.3 | -91.4 | 3,407 |
| 108 | 1415.3 | Mistral Large 3 | Mistral AI | 1411.9 to 1418.7 | -91.5 | 50,153 |
| 109 | 1415.3 | Qwen3 Vl 235b A22b Instruct | Alibaba | 1408.8 to 1421.8 | -91.5 | 11,498 |
| 110 | 1415.3 | DeepSeek V3.1 Terminus | DeepSeek | 1405.6 to 1424.9 | -91.5 | 3,690 |
| 111 | 1413.7 | GPT 4.1 2025 04 14 | OpenAI | 1410.0 to 1417.4 | -93.1 | 50,929 |
| 112 | 1412.5 | Claude Opus 4 20250514 | Anthropic | 1408.2 to 1416.8 | -94.3 | 44,168 |
| 113 | 1412.2 | Claude Haiku 4 5 20251001 | Anthropic | 1409.5 to 1414.9 | -94.6 | 107,630 |
| 114 | 1412.0 | Hunyuan Hy3 Preview | Tencent | 1404.5 to 1419.6 | -94.8 | 6,637 |
| 115 | 1411.5 | Grok 3 Preview 02 24 | xAI | 1407.2 to 1415.9 | -95.3 | 32,894 |
| 116 | 1410.9 | Glm 4.5 | Z.ai | 1406.0 to 1415.8 | -95.9 | 24,285 |
| 117 | 1410.3 | Gemini 2.5 Flash | 1407.8 to 1412.7 | -96.5 | 124,312 | |
| 118 | 1409.6 | Grok 4 0709 | xAI | 1405.7 to 1413.5 | -97.2 | 41,345 |
| 119 | 1409.5 | Mistral Medium 2508 | Mistral AI | 1406.9 to 1412.2 | -97.3 | 93,809 |
| 120 | 1408.7 | Qwen3.5 27b | Alibaba | 1404.3 to 1413.2 | -98.1 | 27,308 |
| 121 | 1404.2 | Gemini 2.5 Flash Preview 09 2025 | 1400.2 to 1408.3 | -102.6 | 32,881 | |
| 122 | 1404.0 | Grok 4 Fast Reasoning | xAI | 1399.0 to 1409.0 | -102.8 | 18,705 |
| 123 | 1403.4 | GPT 5.4 Nano High | OpenAI | 1399.4 to 1407.3 | -103.4 | 55,860 |
| 124 | 1403.0 | Qwen3 235b A22b No Thinking | Alibaba | 1398.5 to 1407.5 | -103.8 | 38,175 |
| 125 | 1402.0 | O1 2024 12 17 | OpenAI | 1397.6 to 1406.5 | -104.8 | 27,807 |
| 126 | 1401.3 | Qwen3 Next 80b A3b Instruct | Alibaba | 1396.5 to 1406.1 | -105.5 | 22,853 |
| 127 | 1401.2 | Longcat Flash Chat | Meituan | 1394.9 to 1407.6 | -105.6 | 11,384 |
| 128 | 1399.2 | Claude Sonnet 4 20250514 Thinking 32k | Anthropic | 1394.8 to 1403.6 | -107.6 | 35,065 |
| 129 | 1399.0 | Qwen3 235b A22b Thinking 2507 | Alibaba | 1392.5 to 1405.6 | -107.8 | 8,985 |
| 130 | 1398.1 | DeepSeek R1 | DeepSeek | 1393.2 to 1403.0 | -108.7 | 18,524 |
| 131 | 1397.3 | Qwen3.5 Flash | Alibaba | 1393.4 to 1401.1 | -109.5 | 55,880 |
| 132 | 1395.7 | Qwen3.5 35b A3b | Alibaba | 1391.4 to 1400.1 | -111.1 | 29,157 |
| 133 | 1395.5 | DeepSeek V3 0324 | DeepSeek | 1391.6 to 1399.4 | -111.3 | 45,480 |
| 134 | 1395.4 | Qwen3 Vl 235b A22b Thinking | Alibaba | 1388.6 to 1402.2 | -111.4 | 7,937 |
| 135 | 1395.2 | Hunyuan Vision 1.5 Thinking | Tencent | 1382.9 to 1407.4 | -111.6 | 2,217 |
| 136 | 1394.9 | Step 3.5 Flash | StepFun | 1391.2 to 1398.6 | -111.9 | 55,180 |
| 137 | 1394.3 | Amazon Nova Experimental Chat 12 10 | Amazon | 1384.7 to 1403.8 | -112.5 | 3,677 |
| 138 | 1392.8 | Mimo V2 Flash (non Thinking) | Xiaomi | 1389.2 to 1396.4 | -114.0 | 46,546 |
| 139 | 1390.6 | Minimax M2.5 | MiniMax | 1386.6 to 1394.6 | -116.2 | 41,092 |
| 140 | 1390.1 | GPT 5 Mini High | OpenAI | 1385.4 to 1394.7 | -116.7 | 27,001 |
| 141 | 1389.9 | O4 Mini 2025 04 16 | OpenAI | 1386.0 to 1393.9 | -116.9 | 45,414 |
| 142 | 1389.2 | Claude Sonnet 4 20250514 | Anthropic | 1384.8 to 1393.5 | -117.6 | 40,277 |
| 143 | 1388.3 | O1 Preview | OpenAI | 1383.3 to 1393.3 | -118.5 | 31,122 |
| 144 | 1387.5 | Qwen3 Coder 480b A35b Instruct | Alibaba | 1382.6 to 1392.5 | -119.3 | 25,694 |
| 145 | 1387.3 | Claude 3 7 Sonnet 20250219 Thinking 32k | Anthropic | 1383.1 to 1391.5 | -119.5 | 38,814 |
| 146 | 1387.1 | Mimo V2 Flash (thinking) | Xiaomi | 1380.9 to 1393.3 | -119.7 | 10,937 |
| 147 | 1386.9 | Hunyuan T1 20250711 | Tencent | 1378.2 to 1395.5 | -119.9 | 4,698 |
| 148 | 1386.8 | Mistral Medium 2505 | Mistral AI | 1382.1 to 1391.5 | -120.0 | 33,193 |
| 149 | 1384.2 | Minimax M2.1 Preview | MiniMax | 1379.0 to 1389.4 | -122.6 | 17,072 |
| 150 | 1383.1 | Qwen3 30b A3b Instruct 2507 | Alibaba | 1378.2 to 1388.0 | -123.7 | 23,715 |
| 151 | 1382.8 | GPT 4.1 Mini 2025 04 14 | OpenAI | 1378.5 to 1387.1 | -124.0 | 39,298 |
| 152 | 1382.2 | Hunyuan Turbos 20250416 | Tencent | 1375.8 to 1388.7 | -124.6 | 10,726 |
| 153 | 1379.6 | Gemini 2.5 Flash Lite Preview 09 2025 No Thinking | 1376.2 to 1383.1 | -127.2 | 47,176 | |
| 155 | 1377.2 | Glm 4.6v | Z.ai | 1365.9 to 1388.5 | -129.6 | 2,801 |
| 156 | 1374.8 | Qwen3 235b A22b | Alibaba | 1370.1 to 1379.5 | -132.0 | 26,257 |
| 157 | 1374.3 | Gemini 2.5 Flash Lite Preview 06 17 Thinking | 1369.8 to 1378.9 | -132.5 | 32,864 | |
| 158 | 1374.2 | Qwen2.5 Max | Alibaba | 1370.1 to 1378.3 | -132.6 | 32,613 |
| 159 | 1373.1 | Glm 4.5 Air | Z.ai | 1368.8 to 1377.3 | -133.7 | 31,062 |
| 160 | 1373.1 | Claude 3 5 Sonnet 20241022 | Anthropic | 1369.9 to 1376.2 | -133.7 | 88,322 |
| 161 | 1371.4 | Claude 3 7 Sonnet 20250219 | Anthropic | 1367.4 to 1375.3 | -135.4 | 43,161 |
| 162 | 1369.5 | Qwen3 Next 80b A3b Thinking | Alibaba | 1363.7 to 1375.4 | -137.3 | 13,678 |
| 164 | 1367.9 | Glm 4.7 Flash | Z.ai | 1362.1 to 1373.7 | -138.9 | 11,706 |
| 165 | 1366.2 | Amazon Nova Experimental Chat 11 10 | Amazon | 1361.9 to 1370.6 | -140.6 | 25,322 |
| 166 | 1365.7 | Gemma 3 27b It | 1362.0 to 1369.4 | -141.1 | 47,500 | |
| 167 | 1363.8 | Minimax M1 | MiniMax | 1359.5 to 1368.0 | -143.0 | 35,156 |
| 168 | 1363.5 | O3 Mini High | OpenAI | 1358.2 to 1368.7 | -143.3 | 18,589 |
| 169 | 1362.0 | Grok 3 Mini High | xAI | 1356.6 to 1367.3 | -144.8 | 16,949 |
| 170 | 1361.6 | Nvidia Nemotron 3 Super 120b A12b | NVIDIA | 1354.3 to 1368.8 | -145.2 | 7,542 |
| 171 | 1360.3 | Gemini 2.0 Flash 001 | 1356.5 to 1364.1 | -146.5 | 43,741 | |
| 172 | 1358.5 | DeepSeek V3 | DeepSeek | 1353.7 to 1363.2 | -148.3 | 21,770 |
| 173 | 1357.5 | Mistral Small 2506 | Mistral AI | 1352.3 to 1362.7 | -149.3 | 17,696 |
| 174 | 1356.7 | Grok 3 Mini Beta | xAI | 1351.7 to 1361.7 | -150.1 | 22,688 |
| 176 | 1353.9 | Command A 03 2025 | Cohere | 1350.5 to 1357.4 | -152.9 | 56,224 |
| 177 | 1353.5 | Glm 4.5v | Z.ai | 1345.2 to 1361.9 | -153.3 | 4,955 |
| 178 | 1353.4 | Gemini 2.0 Flash Lite Preview 02 05 | 1349.1 to 1357.7 | -153.4 | 24,955 | |
| 179 | 1352.5 | GPT Oss 120b | OpenAI | 1348.2 to 1356.9 | -154.3 | 30,611 |
| 180 | 1351.0 | Gemini 1.5 Pro 002 | 1347.7 to 1354.4 | -155.8 | 55,606 | |
| 181 | 1349.7 | Amazon Nova Experimental Chat 10 20 | Amazon | 1343.5 to 1355.9 | -157.1 | 11,460 |
| 182 | 1348.9 | Hunyuan Turbos 20250226 | Tencent | 1337.1 to 1360.6 | -157.9 | 2,220 |
| 183 | 1348.3 | Step 3 | StepFun | 1340.9 to 1355.8 | -158.5 | 6,533 |
| 184 | 1347.7 | O3 Mini | OpenAI | 1344.2 to 1351.2 | -159.1 | 57,313 |
| 185 | 1347.7 | Amazon Nova Experimental Chat 10 09 | Amazon | 1336.8 to 1358.5 | -159.1 | 2,824 |
| 186 | 1347.5 | Llama 3.1 Nemotron Ultra 253b V1 | NVIDIA | 1335.8 to 1359.1 | -159.3 | 2,549 |
| 187 | 1347.2 | Qwen3 32b | Alibaba | 1337.7 to 1356.7 | -159.6 | 3,926 |
| 188 | 1346.7 | Mercury 2 | Inception AI | 1336.1 to 1357.3 | -160.1 | 3,118 |
| 189 | 1346.2 | Ling Flash 2.0 | Ant Group | 1339.0 to 1353.5 | -160.6 | 6,998 |
| 190 | 1346.1 | Qwen Plus 0125 | Alibaba | 1337.8 to 1354.4 | -160.7 | 5,819 |
| 191 | 1346.0 | Minimax M2 | MiniMax | 1338.3 to 1353.8 | -160.8 | 6,860 |
| 192 | 1345.8 | GPT 4o 2024 05 13 | OpenAI | 1342.3 to 1349.2 | -161.0 | 112,881 |
| 193 | 1343.2 | Nvidia Llama 3.3 Nemotron Super 49b V1.5 | NVIDIA | 1333.2 to 1353.1 | -163.6 | 3,344 |
| 194 | 1342.8 | Glm 4 Plus 0111 | Z.ai | 1334.4 to 1351.3 | -164.0 | 5,760 |
| 195 | 1342.5 | Claude 3 5 Sonnet 20240620 | Anthropic | 1339.0 to 1345.9 | -164.3 | 82,419 |
| 196 | 1341.9 | Gemma 3 12b It | 1332.4 to 1351.4 | -164.9 | 3,829 | |
| 197 | 1340.7 | Hunyuan Turbo 0110 | Tencent | 1329.1 to 1352.2 | -166.1 | 2,290 |
| 198 | 1337.2 | GPT 5 Nano High | OpenAI | 1330.3 to 1344.1 | -169.6 | 8,260 |
| 199 | 1337.0 | O1 Mini | OpenAI | 1333.4 to 1340.7 | -169.8 | 51,981 |
| 200 | 1336.8 | Nova 2 Lite | Amazon | 1330.6 to 1342.9 | -170.0 | 12,211 |
| 201 | 1336.1 | Qwq 32b | Alibaba | 1331.7 to 1340.5 | -170.7 | 25,376 |
| 202 | 1335.6 | Grok 2 2024 08 13 | xAI | 1331.9 to 1339.3 | -171.2 | 63,498 |
| 203 | 1335.3 | Gemini Advanced 0514 | 1330.2 to 1340.5 | -171.5 | 50,148 | |
| 204 | 1335.1 | GPT 4o 2024 08 06 | OpenAI | 1330.9 to 1339.3 | -171.7 | 45,499 |
| 205 | 1334.8 | Llama 3.1 405b Instruct Bf16 | Meta | 1331.1 to 1338.5 | -172.0 | 41,375 |
| 206 | 1333.9 | Step 2 16k Exp 202412 | StepFun | 1325.3 to 1342.5 | -172.9 | 4,833 |
| 207 | 1333.0 | Llama 3.1 405b Instruct Fp8 | Meta | 1329.4 to 1336.6 | -173.8 | 59,656 |
| 208 | 1330.0 | Olmo 3.1 32b Instruct | Ai2 | 1323.9 to 1336.1 | -176.8 | 12,205 |
| 209 | 1328.8 | Molmo 2 8b | Ai2 | 1307.4 to 1350.1 | -178.0 | 798 |
| 211 | 1328.0 | Llama 3.3 Nemotron 49b Super V1 | NVIDIA | 1315.8 to 1340.1 | -178.8 | 2,218 |
| 212 | 1327.2 | Qwen3 30b A3b | Alibaba | 1322.4 to 1331.9 | -179.6 | 26,469 |
| 213 | 1327.0 | Llama 4 Maverick 17b 128e Instruct | Meta | 1322.8 to 1331.3 | -179.8 | 39,951 |
| 214 | 1326.2 | Hunyuan Large 2025 02 10 | Tencent | 1316.5 to 1335.9 | -180.6 | 3,738 |
| 215 | 1324.0 | GPT 4 Turbo 2024 04 09 | OpenAI | 1320.1 to 1327.9 | -182.8 | 98,114 |
| 216 | 1323.7 | Claude 3 5 Haiku 20241022 | Anthropic | 1320.5 to 1327.0 | -183.1 | 69,934 |
| 217 | 1323.6 | Gemini 1.5 Pro 001 | 1319.6 to 1327.6 | -183.2 | 79,138 | |
| 218 | 1323.4 | DeepSeek V2.5 1210 | DeepSeek | 1315.2 to 1331.7 | -183.4 | 6,795 |
| 219 | 1322.8 | Llama 4 Scout 17b 16e Instruct | Meta | 1318.1 to 1327.5 | -184.0 | 30,273 |
| 220 | 1322.1 | GPT 4.1 Nano 2025 04 14 | OpenAI | 1314.3 to 1329.8 | -184.7 | 6,103 |
| 221 | 1321.4 | Claude 3 Opus 20240229 | Anthropic | 1318.4 to 1324.5 | -185.4 | 194,909 |
| 222 | 1320.6 | Ring Flash 2.0 | Ant Group | 1313.4 to 1327.8 | -186.2 | 7,135 |
| 223 | 1320.2 | Step 1o Turbo 202506 | StepFun | 1313.4 to 1327.0 | -186.6 | 9,032 |
| 224 | 1319.4 | Glm 4 Plus | Z.ai | 1314.4 to 1324.3 | -187.4 | 26,126 |
| 225 | 1318.2 | Llama 3.3 70b Instruct | Meta | 1314.7 to 1321.7 | -188.6 | 54,720 |
| 226 | 1318.0 | Gemma 3n E4b It | 1312.9 to 1323.2 | -188.8 | 22,569 | |
| 227 | 1318.0 | Qwen Max 0919 | Alibaba | 1312.3 to 1323.7 | -188.8 | 16,478 |
| 228 | 1317.7 | GPT 4o Mini 2024 07 18 | OpenAI | 1314.1 to 1321.2 | -189.1 | 68,709 |
| 229 | 1317.2 | GPT Oss 20b | OpenAI | 1310.9 to 1323.6 | -189.6 | 10,620 |
| 230 | 1315.6 | Nvidia Nemotron 3 Nano 30b A3b Bf16 | NVIDIA | 1310.0 to 1321.1 | -191.2 | 15,496 |
| 231 | 1314.9 | Qwen2.5 Plus 1127 | Alibaba | 1308.5 to 1321.2 | -191.9 | 10,187 |
| 233 | 1314.0 | Mistral Large 2407 | Mistral AI | 1310.1 to 1317.9 | -192.8 | 45,459 |
| 234 | 1312.7 | GPT 4 0125 Preview | OpenAI | 1308.6 to 1316.8 | -194.1 | 93,439 |
| 235 | 1312.3 | GPT 4 1106 Preview | OpenAI | 1308.4 to 1316.2 | -194.5 | 100,105 |
| 236 | 1311.1 | Hunyuan Standard 2025 02 10 | Tencent | 1301.5 to 1320.8 | -195.7 | 3,904 |
| 237 | 1309.2 | Gemini 1.5 Flash 002 | 1305.0 to 1313.4 | -197.6 | 34,902 | |
| 238 | 1308.4 | Grok 2 Mini 2024 08 13 | xAI | 1304.7 to 1312.1 | -198.4 | 52,567 |
| 239 | 1307.4 | Granite 4.1 8b | IBM | 1297.3 to 1317.5 | -199.4 | 4,064 |
| 240 | 1307.1 | DeepSeek V2.5 | DeepSeek | 1302.4 to 1311.8 | -199.7 | 24,572 |
| 242 | 1306.1 | Mercury | Inception AI | 1292.3 to 1319.9 | -200.7 | 1,951 |
| 243 | 1305.6 | Olmo 3 32b Think | Ai2 | 1297.4 to 1313.7 | -201.2 | 5,938 |
| 244 | 1305.3 | Mistral Large 2411 | Mistral AI | 1300.8 to 1309.7 | -201.5 | 28,073 |
| 245 | 1304.1 | Magistral Medium 2506 | Mistral AI | 1297.7 to 1310.6 | -202.7 | 11,624 |
| 246 | 1303.3 | Mistral Small 3.1 24b Instruct 2503 | Mistral AI | 1298.8 to 1307.8 | -203.5 | 33,194 |
| 247 | 1303.3 | Gemma 3 4b It | 1294.0 to 1312.6 | -203.5 | 4,171 | |
| 248 | 1302.8 | Qwen2.5 72b Instruct | Alibaba | 1298.7 to 1306.9 | -204.0 | 39,406 |
| 249 | 1298.9 | Llama 3.1 Nemotron 70b Instruct | NVIDIA | 1291.2 to 1306.7 | -207.9 | 7,140 |
| 250 | 1294.0 | Hunyuan Large Vision | Tencent | 1284.9 to 1303.1 | -212.8 | 5,369 |
| 251 | 1293.1 | Llama 3.1 70b Instruct | Meta | 1289.4 to 1296.9 | -213.7 | 55,240 |
| 252 | 1290.0 | Amazon Nova Pro V1.0 | Amazon | 1285.4 to 1294.6 | -216.8 | 24,745 |
| 254 | 1288.9 | Gemma 2 27b It | 1285.5 to 1292.3 | -217.9 | 75,754 | |
| 256 | 1287.0 | Ibm Granite H Small | IBM | 1278.6 to 1295.4 | -219.8 | 5,682 |
| 257 | 1286.7 | GPT 4 0314 | OpenAI | 1282.0 to 1291.5 | -220.1 | 54,173 |
| 258 | 1286.2 | Gemini 1.5 Flash 001 | 1281.7 to 1290.7 | -220.6 | 62,833 | |
| 259 | 1285.9 | Llama 3.1 Nemotron 51b Instruct | NVIDIA | 1276.0 to 1295.9 | -220.9 | 3,749 |
| 260 | 1285.9 | Llama 3.1 Tulu 3 70b | Ai2 | 1275.5 to 1296.4 | -220.9 | 2,846 |
| 261 | 1285.0 | Olmo 3.1 32b Think | Ai2 | 1277.8 to 1292.2 | -221.8 | 8,497 |
| 262 | 1280.6 | Claude 3 Sonnet 20240229 | Anthropic | 1276.6 to 1284.6 | -226.2 | 109,284 |
| 264 | 1276.5 | Nemotron 4 340b Instruct | NVIDIA | 1271.2 to 1281.9 | -230.3 | 19,659 |
| 265 | 1275.9 | Llama 3 70b Instruct | Meta | 1272.4 to 1279.5 | -230.9 | 156,876 |
| 266 | 1275.8 | Command R Plus 08 2024 | Cohere | 1269.2 to 1282.4 | -231.0 | 9,866 |
| 267 | 1275.0 | GPT 4 0613 | OpenAI | 1270.9 to 1279.1 | -231.8 | 88,723 |
| 268 | 1274.1 | Mistral Small 24b Instruct 2501 | Mistral AI | 1268.2 to 1280.0 | -232.7 | 14,681 |
| 269 | 1273.0 | Glm 4 0520 | Z.ai | 1266.0 to 1280.0 | -233.8 | 9,788 |
| 271 | 1270.4 | Qwen2.5 Coder 32b Instruct | Alibaba | 1262.2 to 1278.5 | -236.4 | 5,432 |
| 272 | 1266.9 | C4ai Aya Expanse 32b | Cohere | 1262.0 to 1271.8 | -239.9 | 27,124 |
| 273 | 1266.4 | Gemma 2 9b It | 1262.6 to 1270.2 | -240.4 | 54,611 | |
| 274 | 1264.5 | DeepSeek Coder V2 | DeepSeek | 1258.3 to 1270.8 | -242.3 | 15,147 |
| 275 | 1261.1 | Qwen2 72b Instruct | Alibaba | 1256.2 to 1266.1 | -245.7 | 37,325 |
| 276 | 1261.1 | Command R Plus | Cohere | 1256.8 to 1265.4 | -245.7 | 77,554 |
| 277 | 1260.9 | Claude 3 Haiku 20240307 | Anthropic | 1257.1 to 1264.6 | -245.9 | 117,701 |
| 278 | 1260.3 | Amazon Nova Lite V1.0 | Amazon | 1255.2 to 1265.5 | -246.5 | 19,372 |
| 279 | 1258.6 | Gemini 1.5 Flash 8b 001 | 1254.3 to 1262.9 | -248.2 | 35,558 | |
| 280 | 1256.0 | Phi 4 | Microsoft | 1251.4 to 1260.6 | -250.8 | 24,126 |
| 281 | 1251.3 | Olmo 2 0325 32b Instruct | Ai2 | 1240.5 to 1262.1 | -255.5 | 3,334 |
| 282 | 1249.7 | Command R 08 2024 | Cohere | 1243.1 to 1256.3 | -257.1 | 10,140 |
| 283 | 1241.7 | Mistral Large 2402 | Mistral AI | 1237.0 to 1246.4 | -265.1 | 62,436 |
| 284 | 1240.6 | Amazon Nova Micro V1.0 | Amazon | 1235.4 to 1245.7 | -266.2 | 19,364 |
| 286 | 1237.3 | Ministral 8b 2410 | Mistral AI | 1228.2 to 1246.4 | -269.5 | 4,781 |
| 287 | 1235.7 | Gemini Pro Dev Api | 1228.4 to 1243.1 | -271.1 | 18,354 | |
| 288 | 1233.5 | Qwen1.5 110b Chat | Alibaba | 1227.9 to 1239.0 | -273.3 | 26,195 |
| 289 | 1233.2 | Hunyuan Standard 256k | Tencent | 1221.5 to 1245.0 | -273.6 | 2,728 |
| 291 | 1232.7 | Qwen1.5 72b Chat | Alibaba | 1227.4 to 1238.0 | -274.1 | 39,302 |
| 292 | 1228.7 | Mixtral 8x22b Instruct V0.1 | Mistral AI | 1224.2 to 1233.3 | -278.1 | 51,416 |
| 293 | 1226.1 | Command R | Cohere | 1221.3 to 1230.9 | -280.7 | 54,036 |
| 295 | 1224.2 | GPT 3.5 Turbo 0125 | OpenAI | 1219.5 to 1228.9 | -282.6 | 66,207 |
| 296 | 1223.0 | Llama 3 8b Instruct | Meta | 1219.2 to 1226.7 | -283.8 | 104,642 |
| 297 | 1222.8 | C4ai Aya Expanse 8b | Cohere | 1215.8 to 1229.8 | -284.0 | 9,818 |
| 298 | 1222.4 | Gemini Pro | 1210.7 to 1234.2 | -284.4 | 6,390 | |
| 299 | 1222.0 | Mistral Medium | Mistral AI | 1216.5 to 1227.5 | -284.8 | 34,550 |
| 300 | 1220.4 | Llama 3.1 Tulu 3 8b | Ai2 | 1209.7 to 1231.1 | -286.4 | 2,896 |
| 303 | 1211.3 | Llama 3.1 8b Instruct | Meta | 1207.1 to 1215.4 | -295.5 | 49,605 |
| 304 | 1207.8 | Granite 3.1 8b Instruct | IBM | 1196.7 to 1218.9 | -299.0 | 3,090 |
| 305 | 1203.2 | Qwen1.5 32b Chat | Alibaba | 1197.0 to 1209.3 | -303.6 | 21,741 |
| 306 | 1202.7 | GPT 3.5 Turbo 1106 | OpenAI | 1194.0 to 1211.5 | -304.1 | 16,619 |
| 307 | 1199.8 | Gemma 2 2b It | 1195.7 to 1203.9 | -307.0 | 46,616 | |
| 308 | 1197.2 | Phi 3 Medium 4k Instruct | Microsoft | 1192.0 to 1202.4 | -309.6 | 25,055 |
| 309 | 1196.3 | Mixtral 8x7b Instruct V0.1 | Mistral AI | 1192.1 to 1200.6 | -310.5 | 73,503 |
| 312 | 1190.3 | Qwen1.5 14b Chat | Alibaba | 1183.2 to 1197.4 | -316.5 | 17,839 |
| 313 | 1183.9 | Wizardlm 70b | Microsoft | 1174.5 to 1193.4 | -322.9 | 8,214 |
| 314 | 1183.9 | DeepSeek LLM 67b Chat | DeepSeek | 1172.4 to 1195.4 | -322.9 | 4,932 |
| 316 | 1182.0 | Granite 3.0 8b Instruct | IBM | 1173.4 to 1190.7 | -324.8 | 6,638 |
| 319 | 1181.5 | Gemma 1.1 7b It | 1175.5 to 1187.6 | -325.3 | 23,893 | |
| 321 | 1178.3 | Granite 3.1 2b Instruct | IBM | 1167.1 to 1189.5 | -328.5 | 3,188 |
| 326 | 1170.4 | Phi 3 Small 8k Instruct | Microsoft | 1164.4 to 1176.3 | -336.4 | 17,766 |
| 327 | 1170.0 | Llama 2 70b Chat | Meta | 1164.5 to 1175.5 | -336.8 | 38,492 |
| 329 | 1166.2 | Llama 3.2 3b Instruct | Meta | 1158.5 to 1173.9 | -340.6 | 7,936 |
| 331 | 1155.7 | Granite 3.0 2b Instruct | IBM | 1147.3 to 1164.1 | -351.1 | 6,837 |
| 332 | 1154.6 | Qwq 32b Preview | Alibaba | 1143.2 to 1166.1 | -352.2 | 3,231 |
| 333 | 1154.0 | Llama2 70b Steerlm Chat | NVIDIA | 1141.4 to 1166.6 | -352.8 | 3,585 |
| 337 | 1148.7 | Mistral 7b Instruct V0.2 | Mistral AI | 1142.0 to 1155.3 | -358.1 | 19,402 |
| 338 | 1148.6 | Wizardlm 13b | Microsoft | 1139.4 to 1157.8 | -358.2 | 7,044 |
| 340 | 1143.1 | Qwen1.5 7b Chat | Alibaba | 1133.3 to 1153.0 | -363.7 | 4,737 |
| 341 | 1142.4 | Phi 3 Mini 4k Instruct June 2024 | Microsoft | 1135.9 to 1148.8 | -364.4 | 12,297 |
| 342 | 1140.8 | Llama 2 13b Chat | Meta | 1134.1 to 1147.5 | -366.0 | 19,174 |
| 344 | 1138.2 | Qwen 14b Chat | Alibaba | 1127.3 to 1149.1 | -368.6 | 4,964 |
| 345 | 1137.7 | Palm 2 | 1128.3 to 1147.0 | -369.1 | 8,554 | |
| 346 | 1136.8 | Gemma 7b It | 1127.3 to 1146.3 | -370.0 | 8,925 | |
| 347 | 1136.0 | Codellama 34b Instruct | Meta | 1127.1 to 1144.9 | -370.8 | 7,366 |
| 349 | 1128.8 | Phi 3 Mini 128k Instruct | Microsoft | 1121.4 to 1136.3 | -378.0 | 20,685 |
| 350 | 1127.5 | Phi 3 Mini 4k Instruct | Microsoft | 1121.1 to 1133.9 | -379.3 | 20,118 |
| 354 | 1118.5 | Codellama 70b Instruct | Meta | 1100.3 to 1136.6 | -388.3 | 1,143 |
| 355 | 1115.5 | Gemma 1.1 2b It | 1107.8 to 1123.3 | -391.3 | 10,854 | |
| 358 | 1110.5 | Llama 3.2 1b Instruct | Meta | 1102.7 to 1118.4 | -396.3 | 8,045 |
| 359 | 1109.2 | Mistral 7b Instruct | Mistral AI | 1100.0 to 1118.5 | -397.6 | 8,977 |
| 360 | 1107.3 | Llama 2 7b Chat | Meta | 1100.3 to 1114.4 | -399.5 | 14,148 |
| 361 | 1092.6 | Gemma 2b It | 1081.1 to 1104.1 | -414.2 | 4,780 | |
| 362 | 1089.9 | Qwen1.5 4b Chat | Alibaba | 1080.6 to 1099.2 | -416.9 | 7,597 |
| 363 | 1073.2 | Olmo 7b Instruct | Ai2 | 1062.0 to 1084.4 | -433.6 | 6,328 |
| 375 | 973.3 | Llama 13b | Meta | 957.5 to 989.1 | -533.5 | 2,391 |
Provider frontier
The leading model per organization.
| Rank | LMArena rating | Organization | Leading model | Reported interval | Gap to leader | Battles |
|---|---|---|---|---|---|---|
| 01 | 1507.5 |  Anthropic | Claude Fable 5 | 1500.2 to 1514.7 | Leader | 8,817 |
| 02 | 1493.1 |  Meta | Muse Spark 1.1 | 1484.9 to 1501.3 | -14.4 | 5,732 |
| 03 | 1486.3 |  OpenAI | GPT-5.6 Sol X-High | 1476.8 to 1495.8 | -21.2 | 4,113 |
| 04 | 1485.8 |  Google | Gemini 3 Pro | 1481.9 to 1489.6 | -21.7 | 41,283 |
| 05 | 1474.2 |  xAI | Grok 4.20 Beta 1 | 1469.5 to 1478.8 | -33.3 | 26,844 |
| 06 | 1457.5 |  DeepSeek | DeepSeek V4 Pro | 1453.1 to 1462.0 | -50.0 | 43,062 |
| 07 | 1427.3 |  Mistral | Mistral Medium 3.5 | 1420.7 to 1433.9 | -80.2 | 11,017 |
Across the boards
Model Gauntlet mirrors named authorities as separate labeled views, each with its own subject, source, date, and license. These columns are not combined and no model is ranked across them.
Human preference
- 01Claude Fable 51507.5
- 02Muse Spark 1.11493.1
- 03GPT-5.6 Sol X-High1486.3
- 04Gemini 3 Pro1485.8
- 05Grok 4.20 Beta 11474.2
- 06DeepSeek V4 Pro1457.5
- 07Mistral Medium 3.51427.3
LMArena leaderboard dataset, CC-BY-4.0. Published July 16, 2026.
Code editing systems
Aider ranks full model-and-editor-system configurations on polyglot code editing, not bare models. These pass rates are not comparable with LMArena preference scores.
- 01gpt-5 (high)88.0%
- 02gpt-5 (medium)86.7%
- 03o3-pro (high)84.9%
- 04gemini-2.5-pro-preview-06-05 (32k think)83.1%
- 05o3 (high)81.3%
- 06gpt-5 (low)81.3%
- 07grok-4 (high)79.6%
- 08gemini-2.5-pro-preview-06-05 (default think)79.1%
- 09o3 (high) + gpt-4.178.2%
- 10Gemini 2.5 Pro Preview 05-0676.9%
Aider polyglot leaderboard, Aider-AI/aider, Apache-2.0. Source last updated May 22, 2026.
Emotional intelligence
- 01claude-opus-4-71461
- 02claude-sonnet-4-61384
- 03claude-opus-4-61383
- 04HiveLabsAI/hivemind-32b-preview1351
- 05deepseek/deepseek-v4-pro1330
- 06openai/gpt-5.41328
- 07openai/gpt-5.51328
- 08moonshotai/kimi-k2.61319
- 09gpt-5.1-2025-11-131314
- 10gpt-5.21314
EQ-Bench 3 canonical Elo leaderboard, EQ-bench/eqbench3, MIT. Source last updated May 10, 2026.
Category boards
This page is the overall text category. LMArena also publishes per-category arenas, each mirrored on its own board with a frontier race by organization.
Leaderboard questions
Which organization leads this leaderboard right now?
According to the LMArena snapshot published July 16, 2026, Anthropic leads with Claude Fable 5 at 1507.5, 14.4 points ahead of Meta in the style-controlled text category.
Is this a combined ranking across many benchmarks?
No. Both tables report one LMArena rating published July 16, 2026: the model rankings list every resolved model with LMArena's official rank, and the provider table shows the highest-scoring model per organization. Model Gauntlet publishes no composite or consensus score.
Why do some rank numbers skip?
The rank column is LMArena's official rank from the published snapshot. Model Gauntlet lists only rows that resolve to a reviewed model identity, so a skipped number is a source row without a reviewed identity match. 324 of the 375 officially ranked rows resolve in this snapshot.
Why does each score show an interval?
LMArena publishes a reported 95% confidence interval with each rating. Where two intervals overlap, this one snapshot does not separate those ranks decisively, so treat small gaps with caution.
How current are these numbers?
The pinned source snapshot was published July 16, 2026 and stays labeled with that date until a newer snapshot passes human review. Nothing on this page is described as live.
Which boards does Model Gauntlet track?
Model Gauntlet mirrors three named authorities as separate labeled views. The LMArena leaderboard dataset covers human preference and fills the tables above, published July 16, 2026. The Aider polyglot leaderboard covers code editing systems, and EQ-Bench 3 covers emotional intelligence; both appear in the Across the boards section below. Each carries its own source, date, and license, and Model Gauntlet publishes no score that combines them.