LMArena Text: Japanese leaderboard
- models measured
- 128
- lead over #2
- +28 Elo
- median
- 1379
Leader on LMArena Text: Japanese · read October 21
On October 21, 2026, 140 of our catalog's models had a LMArena Text: Japanese result, and Anthropic Claude Fable 5.1 led with 1562, ahead of Anthropic Claude Opus 5.5 at 1534.
Prompts written in Japanese. People ask a question, compare two anonymous answers and vote for the better one; the rating is built from those votes, as LMArena publishes it.
Role in the index
LMArena Text: Japanese is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1Anthropic | 1562 | 282 | ±38.3 | 84.2 | $50.00 | |
| 2 | Claude Opus 5.5Anthropic | 1534 | 179 | ±45.6 | 88.8 | $20.00 | |
| 3 | Claude Opus 5Anthropic | 1520 | 594 | ±26.6 | 83.0 | $25.00 | |
| 4 | Gemini 3.7 FlashGoogle | 1511 | 476 | ±29.1 | 58.0 | $3.75 | |
| 5 | Gemini 3.8 FlashGoogle | 1507 | 762 | ±24.2 | 61.6 | $3.75 | |
| 6 | Claude Fable 5Anthropic | 1504 | 643 | ±25.1 | 79.3 | $50.00 | |
| 7 | GPT-5.5OpenAI | 1499 | 907 | ±21.3 | 71.3 | $30.00 | |
| 8 | Kimi K3Moonshot AI | 1494 | 662 | ±25.2 | 62.3 | $15.00 | |
| 9 | Gemini 3.1 Pro PreviewGoogle | 1491 | 1,582 | ±17.2 | 32.3 | $12.00 | |
| 10 | Claude Opus 4.6Anthropic | 1488 | 950 | ±21.1 | 57.2 | $25.00 | |
| 11 | Gemini 3.5 FlashGoogle | 1486 | 700 | ±23.9 | 39.8 | $9.00 | |
| 12 | GPT-5.4OpenAI | 1485 | 676 | ±24.6 | 64.4 | $15.00 | |
| 13 | Gemini 3.6 FlashGoogle | 1475 | 629 | ±25.4 | 62.5 | $3.75 | |
| 14 | Claude Opus 4.7Anthropic | 1474 | 714 | ±24.2 | 67.5 | $25.00 | |
| 15 | Qwen3.8 Max (0902)Qwen | 1473 | 536 | ±27.9 | 62.4 | $6.00 | |
| 16 | GPT-5.6 SolOpenAI | 1472 | 667 | ±25.1 | 73.5 | $20.00 | |
| 17 | Muse Spark 1.3Meta | 1470 | 323 | ±35.0 | 57.4 | $4.25 | |
| 18 | Gemini 2.5 ProGoogle | 1461 | 2,064 | ±14.6 | 19.1 | $10.00 | |
| 19 | GLM 5.3Z.ai | 1460 | 436 | ±29.4 | 70.6 | $4.40 | |
| 20 | GPT-5.6 TerraOpenAI | 1454 | 739 | ±24.1 | 63.9 | $12.00 | |
| 21 | GPT-5.1OpenAI | 1453 | 459 | ±27.7 | 40.4 | $10.00 | |
| 22 | GLM 5.2Z.ai | 1450 | 835 | ±22.4 | 61.1 | $4.40 | |
| 23 | Muse Spark 1.1Meta | 1450 | 783 | ±23.2 | 70.3 | $4.25 | |
| 24 | DeepSeek V4 Pro 0423DeepSeek | 1446 | 721 | ±23.5 | 38.7 | $2.716 | |
| 25 | GLM 5.3 FlashZ.ai | 1444 | 692 | ±24.6 | 57.7 | $0.50 | |
| 26 | Kimi K2.6Moonshot AI | 1443 | 500 | ±27.9 | 52.1 | $4.00 | |
| 27 | Claude Opus 4.8Anthropic | 1442 | 898 | ±21.8 | 65.0 | $25.00 | |
| 28 | GLM 4.7Z.ai | 1439 | 125 | ±54.3 | 55.1 | $2.20 | |
| 29 | GLM 5.1Z.ai | 1435 | 886 | ±21.8 | 49.2 | $4.40 | |
| 30 | DeepSeek V4.1 FlashDeepSeek | 1432 | 285 | ±36.8 | 62.5 | $0.78 | |
| 31 | Grok 4.5xAI | 1430 | 538 | ±27.3 | 58.6 | $6.00 | |
| 32 | Qwen3.5 397B A17BQwen | 1429 | 1,182 | ±19.0 | Not benchmarked | $2.34 | |
| 33 | T InklingThinking Machines | 1429 | 689 | ±24.2 | 31.5 | $4.05 | |
| 34 | Gemini 3.5 Flash LiteGoogle | 1426 | 586 | ±26.5 | 49.0 | $2.50 | |
| 35 | Kimi K2.5Moonshot AI | 1423 | 850 | ±21.6 | 38.6 | $2.85 | |
| 36 | Claude Sonnet 5Anthropic | 1422 | 767 | ±23.4 | 57.7 | $10.00 | |
| 37 | Claude Sonnet 4.6Anthropic | 1420 | 801 | ±22.3 | 57.3 | $15.00 | |
| 38 | GPT-5.2OpenAI | 1420 | 446 | ±29.5 | 55.9 | $14.00 | |
| 39 | Claude Opus 4.5Anthropic | 1417 | 378 | ±31.6 | 44.0 | $25.00 | |
| 40 | GLM 5Z.ai | 1416 | 275 | ±36.2 | 41.7 | $3.20 | |
| 41 | Mistral Large 4Mistral AI | 1415 | 194 | ±43.1 | 56.4 | $4.18 | |
| 42 | GPT-5.6 LunaOpenAI | 1415 | 649 | ±25.4 | 58.0 | $1.20 | |
| 43 | GLM 4.5Z.ai | 1415 | 776 | ±22.7 | Not benchmarked | $2.20 | |
| 44 | Gemini 3.1 Flash Lite PreviewGoogle | 1413 | 647 | ±24.5 | 22.7 | $1.50 | |
| 45 | Qwen3.7 PlusQwen | 1412 | 679 | ±24.1 | 33.2 | $1.28 | |
| 46 | MiMo-V2.5-ProXiaomi | 1410 | 1,104 | ±20.0 | 31.6 | $0.9152 | |
| 47 | GPT-5OpenAI | 1409 | 524 | ±27.1 | 38.8 | $10.00 | |
| 48 | GPT-6 AstraOpenAI | 1405 | 248 | ±40.0 | 85.1 | $50.00 | |
| 49 | o3OpenAI | 1403 | 1,418 | ±17.4 | Not benchmarked | $8.00 | |
| 50 | T Inkling SmallThinking Machines | 1401 | 569 | ±26.8 | 27.3 | $1.20 | |
| 51 | Hy3Tencent | 1397 | 236 | ±39.4 | Not benchmarked | $0.33 | |
| 52 | Gemini 2.5 FlashGoogle | 1396 | 2,044 | ±14.0 | Not benchmarked | $2.50 | |
| 53 | Qwen3 Next 80B A3B InstructQwen | 1395 | 319 | ±32.9 | Not benchmarked | $1.10 | |
| 54 | GLM 4.6Z.ai | 1393 | 373 | ±29.9 | 8.8 | $2.20 | |
| 55 | R1 0528DeepSeek | 1391 | 620 | ±24.9 | Not benchmarked | $2.18 | |
| 56 | Claude Sonnet 4.5Anthropic | 1390 | 785 | ±21.6 | 39.5 | $15.00 | |
| 57 | Qwen3.6 PlusQwen | 1390 | 554 | ±26.4 | 44.1 | $1.95 | |
| 58 | DeepSeek V4 Flash 0423DeepSeek | 1389 | 653 | ±24.7 | Not benchmarked | $0.28 | |
| 59 | Qwen3.8 27BQwen | 1387 | 461 | ±29.0 | 74.2 | $2.55 | |
| 60 | Qwen3 235B A22B Thinking 2507Qwen | 1386 | 442 | ±29.4 | Not benchmarked | $3.50 | |
| 61 | Qwen3 VL 235B A22B InstructQwen | 1385 | 170 | ±45.0 | Not benchmarked | $1.90 | |
| 62 | Step 5 PreviewStepFun | 1383 | 168 | ±45.4 | 61.1 | $2.70 | |
| 63 | MiniMax M3MiniMax | 1380 | 1,005 | ±20.8 | 39.2 | $1.20 | |
| 64 | Grok 4.3xAI | 1379 | 857 | ±22.1 | 37.4 | $2.50 | |
| 65 | Mistral Medium 3.1Mistral AI | 1379 | 1,128 | ±17.9 | Not benchmarked | $2.00 | |
| 66 | Grok 4.6xAI | 1378 | 546 | ±27.6 | 65.7 | $6.00 | |
| 67 | Claude Opus 4.1Anthropic | 1377 | 675 | ±22.4 | Not benchmarked | $75.00 | |
| 68 | GPT-6 SolOpenAI | 1377 | 262 | ±37.3 | 76.0 | $10.00 | |
| 69 | GPT-5.4 MiniOpenAI | 1375 | 623 | ±25.2 | 35.8 | $4.50 | |
| 70 | DeepSeek V3.2DeepSeek | 1374 | 385 | ±29.4 | 30.0 | $0.88 | |
| 71 | GPT-6 LunaOpenAI | 1369 | 1,242 | ±20.2 | 57.3 | $0.50 | |
| 72 | Qwen3.5-FlashQwen | 1369 | 665 | ±23.9 | 41.1 | $0.26 | |
| 73 | Qwen3.5-122B-A10BQwen | 1367 | 257 | ±36.5 | Not benchmarked | $2.08 | |
| 74 | MiMo-V2.6-FlashXiaomi | 1359 | 253 | ±39.8 | 54.7 | $0.28 | |
| 75 | Grok 4.7xAI | 1359 | 281 | ±37.4 | 66.7 | $6.00 | |
| 76 | Step 3.5 FlashStepFun | 1353 | 679 | ±24.1 | Not benchmarked | $0.30 | |
| 77 | GLM 4.5 AirZ.ai | 1348 | 845 | ±20.8 | Not benchmarked | $1.10 | |
| 78 | o1OpenAI | 1346 | 581 | ±26.0 | Not benchmarked | $60.00 | |
| 79 | Qwen3.5-27BQwen | 1346 | 226 | ±39.9 | Not benchmarked | $1.56 | |
| 80 | GPT-5.4 NanoOpenAI | 1343 | 610 | ±25.6 | 44.8 | $1.25 | |
| 81 | GPT-5 MiniOpenAI | 1341 | 460 | ±28.4 | 34.2 | $2.00 | |
| 82 | Claude Haiku 4.5Anthropic | 1338 | 1,762 | ±16.1 | 35.6 | $5.00 | |
| 83 | Qwen3 30B A3B Instruct 2507Qwen | 1337 | 766 | ±22.1 | Not benchmarked | $0.30 | |
| 84 | DeepSeek V3.2 ExpDeepSeek | 1331 | 148 | ±49.5 | Not benchmarked | $0.41 | |
| 85 | gpt-oss-120bOpenAI | 1331 | 578 | ±25.3 | Not benchmarked | $0.50 | |
| 86 | Claude Opus 4RetiredAnthropic | 1330 | 1,115 | ±18.8 | Not benchmarked | $75.00 | |
| 87 | Qwen3.5-35B-A3BQwen | 1326 | 251 | ±38.8 | Not benchmarked | $1.30 | |
| 88 | R1RetiredDeepSeek | 1324 | 468 | ±25.9 | Not benchmarked | $2.50 | |
| 89 | Solar Pro 4Upstage | 1320 | 404 | ±32.2 | Not benchmarked | $0.60 | |
| 90 | GPT-4.1OpenAI | 1319 | 1,138 | ±18.5 | Not benchmarked | $8.00 | |
| 91 | Mistral Medium 3Mistral AI | 1316 | 916 | ±20.0 | Not benchmarked | $2.00 | |
| 92 | Trinity Large ThinkingArcee AI | 1312 | 282 | ±36.4 | Not benchmarked | $0.80 | |
| 93 | Qwen3 Coder 480B A35BQwen | 1311 | 922 | ±20.4 | Not benchmarked | $1.50 | |
| 94 | o4 MiniOpenAI | 1308 | 1,075 | ±19.1 | Not benchmarked | $4.40 | |
| 95 | MiMo-V2.5Xiaomi | 1306 | 540 | ±27.5 | 50.0 | $0.336 | |
| 96 | Gemini 2.0 FlashRetiredGoogle | 1294 | 1,026 | ±18.9 | Not benchmarked | $0.40 | |
| 97 | Gemini 1.5 ProRetiredGoogle | 1292 | 1,210 | ±17.9 | Not benchmarked | $5.00 | |
| 98 | GPT-4.1 MiniOpenAI | 1290 | 1,000 | ±19.6 | Not benchmarked | $1.60 | |
| 99 | Gemma 3 27BGoogle | 1287 | 1,052 | ±18.9 | Not benchmarked | $0.30 | |
| 100 | o3 MiniOpenAI | 1286 | 374 | ±30.2 | Not benchmarked | $4.40 | |
| 101 | Command ACohere | 1285 | 1,236 | ±17.9 | Not benchmarked | $10.00 | |
| 102 | Qwen3 Next 80B A3B ThinkingQwen | 1280 | 206 | ±41.2 | Not benchmarked | $1.20 | |
| 103 | Nova 2 LiteAmazon | 1273 | 115 | ±58.1 | Not benchmarked | $2.50 | |
| 104 | Nemotron 3 Nano 30B A3BNVIDIA | 1268 | 147 | ±51.6 | Not benchmarked | $0.22 | |
| 105 | Claude 3.7 SonnetRetiredAnthropic | 1260 | 1,009 | ±19.3 | Not benchmarked | $15.00 | |
| 106 | MiniMax M2.7MiniMax | 1260 | 1,143 | ±21.1 | 36.2 | $1.20 | |
| 107 | GPT-4o (2024-05-13)OpenAI | 1257 | 2,674 | ±15.4 | Not benchmarked | $15.00 | |
| 108 | Qwen3 30B A3BQwen | 1254 | 733 | ±22.1 | Not benchmarked | $0.50 | |
| 109 | Gemini 1.5 FlashRetiredGoogle | 1252 | 753 | ±21.9 | Not benchmarked | $0.30 | |
| 110 | Qwen-PlusQwen | 1251 | 206 | ±39.0 | Not benchmarked | $0.78 | |
| 111 | o1-miniRetiredOpenAI | 1245 | 1,093 | ±18.6 | Not benchmarked | $4.40 | |
| 112 | gpt-oss-20bOpenAI | 1244 | 246 | ±38.5 | Not benchmarked | $0.15 | |
| 113 | GPT-4o (2024-08-06)OpenAI | 1234 | 1,104 | ±19.0 | Not benchmarked | $10.00 | |
| 114 | Claude 3.5 SonnetRetiredAnthropic | 1234 | 2,040 | ±14.1 | Not benchmarked | $15.00 | |
| 115 | GPT-5 NanoOpenAI | 1227 | 147 | ±49.5 | Not benchmarked | $0.40 | |
| 116 | MiniMax M1MiniMax | 1217 | 845 | ±22.1 | Not benchmarked | $2.20 | |
| 117 | GPT-4o-mini (2024-07-18)OpenAI | 1216 | 1,793 | ±15.5 | Not benchmarked | $0.60 | |
| 118 | Mistral Small 3.1 24BMistral AI | 1208 | 810 | ±21.9 | Not benchmarked | $0.555 | |
| 119 | Llama 4 ScoutMeta Llama | 1206 | 764 | ±22.3 | Not benchmarked | $0.50 | |
| 120 | Llama 4 MaverickMeta Llama | 1206 | 935 | ±21.2 | Not benchmarked | $1.00 | |
| 121 | Claude 3 OpusRetiredAnthropic | 1204 | 3,650 | ±14.4 | Not benchmarked | $75.00 | |
| 122 | Nova Pro 1.0Amazon | 1200 | 511 | ±25.9 | Not benchmarked | $3.20 | |
| 123 | GPT-4.1 NanoOpenAI | 1198 | 160 | ±45.1 | Not benchmarked | $0.40 | |
| 124 | GPT-4 TurboOpenAI | 1194 | 2,112 | ±16.6 | Not benchmarked | $30.00 | |
| 125 | Qwen2.5 72B InstructQwen | 1180 | 788 | ±21.3 | Not benchmarked | $0.40 | |
| 126 | Gemma 2 27BGoogle | 1175 | 1,808 | ±15.9 | Not benchmarked | $0.65 | |
| 127 | Claude 3.5 HaikuRetiredAnthropic | 1175 | 1,546 | ±16.3 | Not benchmarked | $4.00 | |
| 128 | MiniMax M2.5MiniMax | 1171 | 368 | ±37.9 | 37.2 | $1.20 | |
| 129 | Mistral LargeMistral AI | 1170 | 629 | ±22.8 | Not benchmarked | $6.00 | |
| 130 | Command R (08-2024)Cohere | 1166 | 210 | ±38.0 | Not benchmarked | $0.60 | |
| 131 | Phi 4Microsoft | 1158 | 500 | ±26.6 | Not benchmarked | $0.14 | |
| 132 | Nova Micro 1.0Amazon | 1154 | 443 | ±27.6 | Not benchmarked | $0.14 | |
| 133 | Nova Lite 1.0Amazon | 1153 | 403 | ±28.0 | Not benchmarked | $0.24 | |
| 134 | Llama 3.3 70B InstructMeta Llama | 1150 | 1,171 | ±18.6 | Not benchmarked | $0.72 | |
| 135 | GPT-4OpenAI | 1137 | 619 | ±26.0 | Not benchmarked | $60.00 | |
| 136 | Llama 3.1 70B InstructMeta Llama | 1132 | 1,428 | ±17.3 | Not benchmarked | $0.56 | |
| 137 | Claude 3 SonnetRetiredAnthropic | 1131 | 1,753 | ±18.3 | Not benchmarked | $15.00 | |
| 138 | Mistral Small 3Mistral AI | 1111 | 269 | ±36.2 | Not benchmarked | $0.08 | |
| 139 | Claude 3 HaikuRetiredAnthropic | 1102 | 2,063 | ±16.9 | Not benchmarked | $1.25 | |
| 140 | Llama 3.1 8B InstructMeta Llama | 1061 | 1,234 | ±19.3 | Not benchmarked | $0.08 |
Only models in our catalog are listed, and each appears once with its best published run. 184 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.