LLM Leaderboard · AI Model Rankings
Global LLMs 6-Dim Roundup · one chart to compare them all
Covering the top 100 LLMs by Arena Elo, scored across six dimensions: battle popularity, coding, vision, intelligence index, knowledge and reasoning. Click column headers to sort; filter byopen-source / domesticfilter.
100
Models tracked
Arena Elo Top 100
39
Chinese models
Zhipu / Alibaba / DeepSeek / Moonshot…
33
Open-source / open weights
MIT · Apache 2.0 and more
Claude Fable 5.1
Current #1 · Anthropic
Arena Elo 1520 · AAII 53
Arena Elo · Top 10
Human blind-battle scores · bar length = strength · domestic models in orangeOverall Ranking · Top 100
Sorted by Arena Elo by default · click a column header to re-sort| # | Models | Arena Elo | Coding | Vision | Intelligence | MMLU-Pro | ARC-AGI |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 Anthropic · Proprietary | 1520 | 1570 | 1315 | 53 | 92.4 | 90.0 |
| 2 | GPT-6 Astra OpenAI · Proprietary | 1520 | 1568 | 1316 | 53 | 92.0 | 95.0 |
| 3 | Claude Opus 5 Anthropic · Proprietary | 1511 | 1566 | 1314 | 51 | 91.6 | 90.4 |
| 4 | Claude Fable 5 Anthropic · Proprietary | 1510 | 1566 | 1312 | 50 | 91.5 | 89.2 |
| 5 | GPT-5.6 Sol OpenAI · Proprietary | 1509 | 1567 | 1312 | 47 | 90.2 | 92.5 |
| 6 | Grok-4.6 xAI · Proprietary | 1507 | 1563 | 1310 | 47 | 89.6 | 75.0 |
| 7 | Muse Spark 1.3 Meta · Proprietary | 1507 | 1562 | 1310 | 47 | 89.0 | — |
| 8 | Kimi-K3 ✅ Moonshot · Kimi K3 开源国产 | 1506 | 1562 | 1311 | 46 | 89.3 | — |
| 9 | Claude Opus 4.8 Thinking Anthropic · Proprietary | 1506 | 1562 | 1310 | 45 | 90.1 | 78.0 |
| 10 | Qwen3.8-Max ✅ Alibaba · Qwen3 开源国产 | 1506 | 1560 | 1312 | 45 | 89.8 | — |
| 11 | GPT-5.6 Terra OpenAI · Proprietary | 1505 | 1562 | 1310 | 44 | 89.6 | 83.9 |
| 12 | Gemini-3.8-Flash Google · Proprietary | 1505 | 1561 | 1312 | 45 | 90.2 | — |
| 13 | GPT-5.5-high OpenAI · Proprietary | 1505 | 1561 | 1311 | 44 | 89.6 | 85.0 |
| 14 | GLM-5.3 ✅ Z.ai · MIT 开源国产 | 1505 | 1560 | — | 45 | 87.8 | — |
| 15 | Claude Opus 4.7 Thinking Anthropic · Proprietary | 1504 | 1559 | 1309 | 43 | 90.0 | 75.8 |
| 16 | Grok-4.5 xAI · Proprietary | 1504 | 1556 | 1305 | 43 | 89.5 | — |
| 17 | Gemini-3.1-Pro Google · Proprietary | 1504 | 1531 | 1309 | 43 | 91.0 | 77.1 |
| 18 | Claude Opus 4.8 Anthropic · Proprietary | 1503 | 1558 | 1306 | 42 | 90.0 | 62.2 |
| 19 | DeepSeek-V4.1-Flash DeepSeek · MIT 开源国产 | 1503 | 1557 | — | 43 | 87.4 | — |
| 20 | Gemini-3.7-Flash Google · Proprietary | 1503 | 1556 | 1303 | 43 | 90.0 | 80.6 |
| 21 | Claude Opus 4.7 Anthropic · Proprietary | 1502 | 1554 | 1300 | 41 | 89.9 | 62.1 |
| 22 | DeepSeek-V4-Pro DeepSeek · MIT 开源国产 | 1502 | 1550 | — | 41 | 87.5 | — |
| 23 | Claude Opus 4.6 Thinking Anthropic · Proprietary | 1502 | 1545 | 1304 | 41 | 89.7 | 69.2 |
| 24 | Muse Spark 1.2 Meta · Proprietary | 1502 | 1540 | 1300 | 43 | 88.8 | — |
| 25 | Gemini-3.6-Flash Google · Proprietary | 1501 | 1535 | 1301 | 41 | 90.0 | 72.1 |
| 26 | GPT-5.4-high OpenAI · Proprietary | 1496 | 1538 | 1290 | 41 | 88.6 | 74.0 |
| 27 | Muse Spark 1.1 Meta · Proprietary | 1496 | 1521 | 1298 | 41 | 88.7 | — |
| 28 | Grok-4.20 xAI · Proprietary | 1495 | 1518 | 1279 | 41 | 88.6 | 65.1 |
| 29 | Gemini-3-Pro Google · Proprietary | 1492 | 1501 | 1308 | 41 | 90.0 | 33.6 |
| 30 | Claude Opus 4.6 Anthropic · Proprietary | 1490 | 1535 | 1298 | 40 | 89.5 | 64.6 |
| 31 | GLM-5.2 Z.ai · MIT 开源国产 | 1488 | 1525 | — | 41 | 87.5 | 22.8 |
| 32 | Qwen3.7-Max Alibaba · Proprietary 国产 | 1486 | 1505 | 1289 | 41 | 89.6 | — |
| 33 | Claude Sonnet 5 Thinking Anthropic · Proprietary | 1485 | 1520 | 1285 | 41 | 89.0 | — |
| 34 | Muse Spark Meta · Proprietary | 1484 | 1493 | 1294 | 40 | 87.3 | — |
| 35 | Grok-4.1-Thinking xAI · Proprietary | 1482 | 1483 | — | 39 | 88.2 | 26.0 |
| 36 | ERNIE-5.1 Baidu · Proprietary 国产 | 1475 | 1495 | — | 40 | 87.1 | — |
| 37 | Claude Opus 4.5 (thinking-32k) Anthropic · Proprietary | 1469 | 1510 | — | 39 | 89.5 | 30.6 |
| 38 | Claude Sonnet 4.6 Thinking Anthropic · Proprietary | 1468 | 1511 | 1278 | 40 | 88.0 | 60.4 |
| 39 | GLM-5.1 Z.ai · MIT 开源国产 | 1467 | 1506 | — | 40 | 87.1 | 5.1 |
| 40 | Seed2.0 Pro ByteDance · Proprietary 国产 | 1466 | 1495 | 1288 | 39 | 87.8 | — |
| 41 | Kimi-K2.6-Thinking Moonshot · Modified MIT 开源国产 | 1466 | 1493 | 1286 | 40 | 87.3 | — |
| 42 | Qwen3.5-Max Alibaba · Proprietary 国产 | 1466 | 1493 | — | 39 | 87.8 | — |
| 43 | MiMo-V2.5-Pro Xiaomi · MIT 开源国产 | 1466 | 1485 | — | 39 | 87.0 | — |
| 44 | GPT-5.2-high OpenAI · Proprietary | 1465 | 1470 | 1280 | 41 | 87.5 | 52.9 |
| 45 | Gemini-3-Flash Google · Proprietary | 1465 | 1469 | 1292 | 40 | 89.0 | 31.1 |
| 46 | GPT-5.4 OpenAI · Proprietary | 1465 | 1468 | 1275 | 39 | 88.4 | 29.2 |
| 47 | GPT-5.1-high OpenAI · Proprietary | 1464 | 1466 | 1250 | 39 | 87.1 | 17.6 |
| 48 | GPT-5.2 OpenAI · Proprietary | 1464 | 1465 | 1248 | 37 | 87.4 | 26.7 |
| 49 | Grok-4.1 xAI · Proprietary | 1463 | 1463 | — | 37 | 88.0 | — |
| 50 | Claude Opus 4.5 Anthropic · Proprietary | 1462 | 1496 | — | 31 | 88.8 | 7.8 |
| 51 | GPT-5.4-mini-high OpenAI · Proprietary | 1461 | 1461 | — | 37 | 87.0 | 18.9 |
| 52 | Claude Sonnet 4.6 Anthropic · Proprietary | 1460 | 1500 | 1277 | 35 | 87.3 | 37.6 |
| 53 | Gemini-2.5-Pro Google · Proprietary | 1459 | 1465 | 1266 | 34 | 86.2 | 4.9 |
| 54 | ERNIE-5.0 Baidu · Proprietary 国产 | 1458 | 1461 | 1251 | 35 | 86.0 | — |
| 55 | Qwen3.6-Plus Alibaba · Proprietary 国产 | 1456 | 1482 | — | 39 | 88.5 | — |
| 56 | Minimax-M3 MiniMax · Non-commercial 国产 | 1452 | 1485 | — | 39 | 87.2 | — |
| 57 | GLM-5 Z.ai · MIT 开源国产 | 1452 | 1461 | — | 39 | 87.0 | 5.0 |
| 58 | Kimi-K2.5-Thinking Moonshot · Modified MIT 开源国产 | 1451 | 1480 | 1271 | 38 | 87.1 | 11.8 |
| 59 | DeepSeek-V4-Flash DeepSeek · MIT 开源国产 | 1451 | 1466 | — | 39 | 86.2 | 61.4 |
| 60 | GPT-5.6 Luna OpenAI · Proprietary | 1451 | 1465 | — | 39 | 86.1 | 59.5 |
| 61 | Qwen3.8-27B Alibaba · Apache 2.0 开源国产 | 1450 | 1465 | — | 39 | 86.3 | — |
| 62 | Qwen3.5-397B-A17B Alibaba · Apache 2.0 开源国产 | 1450 | 1463 | — | 38 | 87.8 | 5.0 |
| 63 | Gemma-4-31B-it Google · Apache 2.0 开源 | 1449 | 1462 | — | 34 | 85.2 | — |
| 64 | GLM-4.7 Z.ai · MIT 开源国产 | 1445 | 1460 | — | 37 | 85.6 | — |
| 65 | GPT-5-high OpenAI · Proprietary | 1444 | 1460 | 1232 | 35 | 87.1 | 9.9 |
| 66 | Qwen3-Max Alibaba · Proprietary 国产 | 1443 | 1468 | — | 34 | 85.3 | — |
| 67 | Gemma-4-26B-A4B-it Google · Apache 2.0 开源 | 1443 | 1460 | — | 28 | 82.6 | — |
| 68 | Grok-4 xAI · Proprietary | 1442 | 1453 | 1221 | 35 | 86.6 | 16.0 |
| 69 | GLM-4.6 Z.ai · MIT 开源国产 | 1441 | 1458 | — | 28 | 83.5 | — |
| 70 | GPT-5.1 OpenAI · Proprietary | 1440 | 1450 | 1243 | 35 | 87.0 | 7.5 |
| 71 | Kimi-K2-Thinking Moonshot · Modified MIT 开源国产 | 1438 | 1450 | — | 37 | 84.8 | — |
| 72 | Claude Sonnet 4.5 (thinking-32k) Anthropic · Proprietary | 1432 | 1485 | — | 32 | 87.5 | 13.6 |
| 73 | MiMo-V2-Pro Xiaomi · Proprietary 国产 | 1431 | 1441 | — | 38 | 86.8 | — |
| 74 | GLM-4.5 Z.ai · MIT 开源国产 | 1430 | 1448 | — | 26 | 83.5 | — |
| 75 | Qwen3-VL-235B-A22B-Instruct Alibaba · Apache 2.0 开源国产 | 1429 | 1457 | 1246 | 22 | 82.8 | — |
| 76 | Mistral Large 3 Mistral · Apache 2.0 开源 | 1428 | 1450 | — | 16 | 81.0 | — |
| 77 | ChatGPT-4o-latest (2025-03-26) OpenAI · Proprietary | 1427 | 1434 | 1242 | 15 | 80.3 | — |
| 78 | DeepSeek-R1-0528 DeepSeek · MIT 开源国产 | 1426 | 1436 | — | 29 | 84.9 | 1.3 |
| 79 | Claude Opus 4.1 (thinking-16k) Anthropic · Proprietary | 1425 | 1475 | — | 30 | 87.8 | — |
| 80 | o3-2025-04-16 OpenAI · Proprietary | 1424 | 1441 | 1218 | 35 | 85.3 | 6.5 |
| 81 | Grok-3-Preview-02-24 xAI · Proprietary | 1423 | 1439 | — | 18 | 79.9 | — |
| 82 | Hunyuan-Hy3 Tencent · Tencent 开源国产 | 1422 | 1448 | — | 35 | 86.2 | — |
| 83 | DeepSeek-V3.2-Thinking DeepSeek · MIT 开源国产 | 1422 | 1438 | — | 35 | 86.2 | 4.0 |
| 84 | Gemini-3.1-Flash-Lite Google · Proprietary | 1421 | 1402 | 1230 | 31 | 86.2 | — |
| 85 | Claude Sonnet 4.5 Anthropic · Proprietary | 1420 | 1464 | — | 22 | 86.0 | — |
| 86 | LongCat-Flash-Chat Meituan · MIT 开源国产 | 1420 | 1461 | — | 22 | 82.7 | — |
| 87 | Claude Opus 4.1 Anthropic · Proprietary | 1419 | 1465 | — | 21 | 87.3 | — |
| 88 | Grok-4-Fast xAI · Proprietary | 1419 | 1441 | — | 33 | 85.0 | — |
| 89 | Qwen3-235B-A22B-Instruct-2507 Alibaba · Apache 2.0 开源国产 | 1418 | 1457 | — | 22 | 82.8 | 1.3 |
| 90 | Nemotron-3-Ultra-550B-A55B Nvidia · Nvidia Open 开源 | 1418 | 1452 | — | 39 | 86.8 | — |
| 91 | DeepSeek-V3.1-Thinking DeepSeek · MIT 开源国产 | 1418 | 1437 | — | 30 | 85.1 | — |
| 92 | DeepSeek-V3.2 DeepSeek · MIT 开源国产 | 1418 | 1431 | — | 25 | 83.7 | — |
| 93 | Qwen3-Next-80B-A3B-Instruct Alibaba · Apache 2.0 开源国产 | 1417 | 1456 | — | 29 | 82.4 | — |
| 94 | Amazon-Nova-Chat-11-10 Amazon · Proprietary | 1417 | 1431 | — | 33 | 83.0 | — |
| 95 | DeepSeek-V3.1 DeepSeek · MIT 开源国产 | 1417 | 1430 | — | 21 | 83.3 | — |
| 96 | Minimax-M2.7 MiniMax · Non-commercial 国产 | 1416 | 1448 | — | 38 | 87.1 | — |
| 97 | Qwen3-235B-A22B-Thinking-2507 Alibaba · Apache 2.0 开源国产 | 1416 | 1442 | — | 33 | 84.3 | — |
| 98 | GPT-4.5-Preview OpenAI · Proprietary | 1415 | 1419 | 1190 | 16 | 81.0 | 0.8 |
| 99 | GPT-5-chat OpenAI · Proprietary | 1413 | 1421 | 1238 | 35 | 86.6 | 7.5 |
| 100 | Gemini-2.5-Flash Google · Proprietary | 1412 | 1420 | 1235 | 28 | 83.2 | 2.5 |
Score notes:Arena Elo is a million-scale human blind-battle score (aggregated by openlm.ai; higher = stronger);AAII is the Artificial Analysis intelligence index (aggregating MMLU-Pro, GPQA, HLE and 7 more benchmarks);ARC-AGI is a fluid-intelligence reasoning benchmark (out of 100); "—" means no data for that dimension. Data source: openlm.ai / lmmarketcap, for model-selection reference only.