Leaderboard
AI model leaderboard
Which models people prefer when they compare two answers side by side, from LMArena’s open votes. It measures preference, not correctness, and the ± range shows how sure the ranking is. Prices from our API price table.
General chat, writing and reasoning, voted by people side by side. Leaderboard of 2026-11-09. Checked .
| # | Model | Score | Votes | Licence | API $/M in · out |
|---|---|---|---|---|---|
| 1 | Gemini 4 Carbon (High) google | 1538 ±10 | 4,282 | Proprietary | — |
| 2 | Gemini 4 Argon (High) google | 1530 ±5 | 19,787 | Proprietary | — |
| 3 | Claude Opus 5.5 (High) anthropic | 1513 ±6 | 15,249 | Proprietary | — |
| 4 | Claude Fable 5.1 (Max) anthropic | 1512 ±6 | 20,087 | Proprietary | — |
| 5 | Claude Opus 5 (Max) anthropic | 1506 ±5 | 35,710 | Proprietary | — |
| 6 | Muse Wonder (Max) meta | 1503 ±6 | 12,032 | Proprietary | — |
| 7 | Claude Opus 5 (High) anthropic | 1502 ±4 | 73,697 | Proprietary | — |
| 8 | Claude Opus 4.6 (High) anthropic | 1502 ±3 | 84,424 | Proprietary | — |
| 9 | Gemini 3.8 Flash (High) google | 1500 ±4 | 61,143 | Proprietary | — |
| 10 | Claude Opus 4.6 anthropic | 1497 ±3 | 89,228 | Proprietary | $5 · $25 |
| 11 | MiMo-V2.6-Pro xiaomi | 1492 ±9 | 4,066 | MIT | — |
| 12 | Claude Fable 5 (High) anthropic | 1492 ±4 | 45,524 | Proprietary | — |
| 13 | Muse Spark 1.3 (Max) meta | 1491 ±5 | 26,111 | Proprietary | — |
| 14 | Claude Opus 4.7 (High) anthropic | 1490 ±4 | 64,308 | Proprietary | — |
| 15 | Gemini 3.7 Flash (High) google | 1489 ±5 | 33,300 | Proprietary | — |
| 16 | Qwen3.8-Max alibaba | 1485 ±5 | 36,324 | Proprietary | $2 · $6 |
| 17 | Muse Spark 1.2 (xHigh) meta | 1484 ±6 | 15,004 | Proprietary | — |
| 18 | Claude Opus 4.7 anthropic | 1483 ±4 | 65,369 | Proprietary | $5 · $25 |
| 19 | Gemini 3.5 Flash (High) google | 1483 ±4 | 47,642 | Proprietary | — |
| 20 | Gemini 3.1 Pro Preview google | 1480 ±3 | 132,296 | Proprietary | $2 · $12 |
| 21 | Gemini 3 Pro google | 1480 ±4 | 41,829 | Proprietary | — |
| 22 | Gemini 3.6 Flash (High) google | 1479 ±4 | 35,405 | Proprietary | — |
| 23 | Muse Spark 1.1 meta | 1479 ±4 | 45,562 | Proprietary | — |
| 24 | Gemini 3.5 Flash (Medium) google | 1478 ±4 | 46,377 | Proprietary | — |
| 25 | GLM-5.3 (Max) zai | 1476 ±5 | 31,017 | MIT | — |
| 26 | Qwen3.7-Max-Preview alibaba | 1475 ±10 | 3,904 | Proprietary | $2.5 · $7.5 |
| 27 | Kimi K3 (Max) moonshot | 1474 ±4 | 40,867 | Kimi K3 license | — |
| 28 | Muse Spark meta | 1474 ±6 | 14,009 | Proprietary | — |
| 29 | GPT-5.5 (High) openai | 1471 ±4 | 68,359 | Proprietary | — |
| 30 | Qwen3.5-Max-Preview alibaba | 1471 ±5 | 22,419 | Proprietary | — |
| 31 | GLM-5.3-Flash zai | 1471 ±4 | 50,189 | MIT | $0.15 · $0.5 |
| 32 | GLM-5.2 (Max) zai | 1469 ±4 | 54,173 | MIT | — |
| 33 | GPT-5.4 (High) openai | 1469 ±4 | 64,103 | Proprietary | — |
| 34 | Claude Sonnet 5.5 (xHigh) anthropic | 1469 ±6 | 13,821 | Proprietary | — |
| 35 | ERNIE 5.1 baidu | 1468 ±5 | 39,029 | Proprietary | — |
| 36 | GPT-5.5 openai | 1466 ±4 | 69,937 | Proprietary | $5 · $30 |
| 37 | Gemini 3 Flash google | 1466 ±4 | 31,133 | Proprietary | $0.5 · $3 |
| 38 | MiMo-V2.5-Pro xiaomi | 1464 ±4 | 76,793 | MIT | — |
| 39 | DeepSeek-V4.1-Flash-20260910 (Max) deepseek | 1463 ±6 | 13,580 | MIT | — |
| 40 | GLM-5.1 zai | 1463 ±4 | 67,074 | MIT | $1.4 · $4.4 |
| 41 | Claude Opus 4.8 (High) anthropic | 1461 ±4 | 63,128 | Proprietary | — |
| 42 | Gemini 2.5 Pro google | 1458 ±2 | 124,789 | Proprietary | $1.25 · $10 |
| 43 | Claude Sonnet 4.6 anthropic | 1458 ±4 | 70,017 | Proprietary | $3 · $15 |
| 44 | Kimi K3.1 Pro (Max) moonshot | 1457 ±11 | 3,443 | Kimi K3.1 Pro License | — |
| 45 | GPT-5.6 Sol (xHigh) openai | 1457 ±4 | 43,994 | Proprietary | — |
| 46 | Kimi K2.6 moonshot | 1455 ±4 | 39,573 | Modified MIT | $0.95 · $4 |
| 47 | Qwen3.7-Plus alibaba | 1454 ±4 | 41,318 | Proprietary | — |
| 48 | Claude Opus 4.8 anthropic | 1454 ±4 | 64,170 | Proprietary | $5 · $25 |
| 49 | GPT-5.4 openai | 1452 ±4 | 67,105 | Proprietary | $2.5 · $15 |
| 50 | DeepSeek-V4-Pro deepseek | 1451 ±4 | 56,906 | MIT | $1.32 · $3.96 |
| 51 | Grok 4.20 Beta (Reasoning, 0309) xai | 1451 ±4 | 65,713 | Proprietary | — |
| 52 | Claude Opus 4.5 (20251101) anthropic | 1451 ±3 | 72,313 | Proprietary | — |
| 53 | Grok 4.20 Multi-Agent Beta (0309) xai | 1451 ±4 | 64,252 | Proprietary | — |
| 54 | MiMo-V2.6-Flash xiaomi | 1450 ±5 | 22,354 | MIT | — |
| 55 | GPT-5.6 Terra (xHigh) openai | 1449 ±4 | 48,825 | Proprietary | — |
| 56 | Dola Seed 2.0 Pro bytedance | 1449 ±3 | 78,230 | Proprietary | — |
| 57 | amazon-nova-experimental-chat (26-02-10) amazon | 1449 ±10 | 3,510 | Proprietary | — |
| 58 | Claude Opus 4.5 (High, 32K, 20251101) anthropic | 1448 ±4 | 37,253 | Proprietary | — |
| 59 | Grok 4.5 xai | 1448 ±4 | 39,155 | Proprietary | $2 · $6 |
| 60 | Qwen3.6-Max-Preview alibaba | 1446 ±8 | 5,386 | Proprietary | — |
| 61 | GLM-5 zai | 1446 ±4 | 29,020 | MIT | $1 · $3.2 |
| 62 | GPT-6.1 Sol (Max) openai | 1446 ±6 | 12,952 | Proprietary | — |
| 63 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) nvidia | 1445 ±7 | 11,399 | OpenMDW-1.1 | — |
| 64 | DeepSeek-V4-Pro Preview (High) deepseek | 1445 ±4 | 54,233 | MIT | — |
| 65 | DeepSeek-V4-Pro (High, 20260813) deepseek | 1445 ±7 | 9,584 | MIT | — |
| 66 | Kimi K2.5 (Thinking) moonshot | 1445 ±3 | 73,872 | Modified MIT | — |
| 67 | Grok 4.20 Beta 1 xai | 1445 ±5 | 27,690 | Proprietary | — |
| 68 | ERNIE 5.0 (0110) baidu | 1444 ±4 | 36,175 | Proprietary | — |
| 69 | Inkling thinky | 1444 ±4 | 50,137 | Apache 2.0 | — |
| 70 | Step 5 Preview stepfun | 1443 ±6 | 12,362 | Proprietary | — |
| 71 | ERNIE-5.0-Preview (1203) baidu | 1443 ±6 | 9,740 | Proprietary | — |
| 72 | GPT-5.1 (High) openai | 1443 ±4 | 41,405 | Proprietary | — |
| 73 | Gemma 4 31B google | 1443 ±7 | 6,130 | Apache 2.0 | — |
| 74 | Gemini 3 Flash (Minimal) google | 1442 ±3 | 90,109 | Proprietary | — |
| 75 | Claude Sonnet 5 (High) anthropic | 1442 ±4 | 53,013 | Proprietary | — |
| 76 | GLM-4.6 zai | 1440 ±4 | 35,652 | MIT | $0.6 · $2.2 |
| 77 | GPT-6 Astra (Max) openai | 1440 ±6 | 18,315 | Proprietary | — |
| 78 | Qwen3-Max-Preview alibaba | 1439 ±5 | 27,301 | Proprietary | — |
| 79 | Hy3 tencent | 1439 ±6 | 15,068 | Apache 2.0 | — |
| 80 | Claude Sonnet 4.5 (20250929) anthropic | 1439 ±3 | 82,119 | Proprietary | — |
| 81 | GPT-5.2 Chat (20260210) openai | 1439 ±4 | 35,708 | Proprietary | — |
| 82 | GLM-5V-Turbo zai | 1438 ±6 | 10,291 | Proprietary | $1.2 · $4 |
| 83 | Qwen3.5-397B-A17B alibaba | 1438 ±3 | 97,678 | Apache 2.0 | — |
| 84 | Grok 4.1 (Thinking) xai | 1437 ±3 | 66,074 | Proprietary | — |
| 85 | MiMo-V2-Pro xiaomi | 1437 ±5 | 25,245 | Proprietary | — |
| 86 | Qwen3.6-Plus alibaba | 1436 ±4 | 47,601 | Proprietary | — |
| 87 | Grok 4.1 xai | 1436 ±3 | 68,194 | Proprietary | — |
| 88 | Qwen3.8-27B alibaba | 1436 ±5 | 32,659 | Apache 2.0 | — |
| 89 | GLM-4.7 zai | 1435 ±6 | 12,167 | MIT | $0.6 · $2.2 |
| 90 | Gemini 3.5 Flash-Lite google | 1435 ±4 | 35,114 | Proprietary | $0.3 · $2.5 |
| 91 | Claude Sonnet 4.5 (High, 32K, 20250929) anthropic | 1435 ±3 | 83,387 | Proprietary | — |
| 92 | Gemma 4 26B A4B google | 1434 ±7 | 6,075 | Apache 2.0 | — |
| 93 | DeepSeek-V4-Flash deepseek | 1432 ±4 | 51,368 | MIT | $0.3 · $1.2 |
| 94 | GPT-5.6 Luna (xHigh) openai | 1431 ±4 | 40,641 | Proprietary | — |
| 95 | GLM-4.5 zai | 1430 ±5 | 23,759 | MIT | $0.6 · $2.2 |
| 96 | Grok 4.6 (High) xai | 1430 ±5 | 34,616 | Proprietary | — |
| 97 | ERNIE-5.0-Preview (1022) baidu | 1430 ±9 | 4,737 | Proprietary | — |
| 98 | Mistral Large 3 mistral | 1429 ±3 | 88,641 | Apache 2.0 | $0.5 · $1.5 |
| 99 | ChatGPT-4o Latest (20250326) openai | 1428 ±3 | 81,663 | Proprietary | — |
| 100 | MiniMax-M3 minimax | 1428 ±4 | 72,963 | MiniMax Community License | — |
Data: LMArena leaderboard dataset, licensed CC BY 4.0. Prices are matched by model name and may be missing for models without a public API.