LMArena Text: Korean leaderboard
- Claude Opus 5 leads
- 1517
- models measured
- 129
- lead over #2
- +1 Elo
- median
- 1375
Leader on LMArena Text: Korean · read October 22
On October 22, 2026, 141 of our catalog's models had a LMArena Text: Korean result, and Anthropic Claude Opus 5 led with 1517, ahead of Anthropic Claude Fable 5.1 at 1517.
Prompts written in Korean. 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: Korean 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 Opus 5Anthropic | 1517 | 648 | ±24.1 | 83.0 | $25.00 | |
| 2 | Claude Fable 5.1Anthropic | 1517 | 238 | ±39.0 | 84.2 | $50.00 | |
| 3 | Muse Spark 1.3Meta | 1490 | 344 | ±32.4 | 57.4 | $4.25 | |
| 4 | Claude Fable 5Anthropic | 1487 | 806 | ±22.4 | 79.3 | $50.00 | |
| 5 | Gemini 3.7 FlashGoogle | 1481 | 524 | ±27.1 | 58.0 | $3.75 | |
| 6 | Gemini 3.8 FlashGoogle | 1464 | 740 | ±23.6 | 61.6 | $3.75 | |
| 7 | Claude Opus 4.6Anthropic | 1464 | 1,552 | ±16.7 | 57.2 | $25.00 | |
| 8 | Claude Opus 4.7Anthropic | 1464 | 1,147 | ±18.6 | 67.5 | $25.00 | |
| 9 | Muse Spark 1.2Meta | 1462 | 196 | ±40.7 | 55.7 | $4.25 | |
| 10 | Qwen3.8 Max (0902)Qwen | 1460 | 608 | ±24.6 | 62.4 | $6.00 | |
| 11 | GPT-5.5OpenAI | 1460 | 1,180 | ±18.3 | 71.3 | $30.00 | |
| 12 | Kimi K3Moonshot AI | 1460 | 717 | ±22.5 | 62.3 | $15.00 | |
| 13 | GLM 5.3Z.ai | 1460 | 404 | ±29.5 | 70.6 | $4.40 | |
| 14 | Muse Spark 1.1Meta | 1459 | 826 | ±21.4 | 70.3 | $4.25 | |
| 15 | Gemini 3.1 Pro PreviewGoogle | 1455 | 2,277 | ±14.3 | 32.3 | $12.00 | |
| 16 | GPT-5.4OpenAI | 1449 | 1,108 | ±19.5 | 64.4 | $15.00 | |
| 17 | DeepSeek V4 Pro 0813DeepSeek | 1449 | 197 | ±40.8 | 56.1 | $1.98 | |
| 18 | Gemini 3.5 FlashGoogle | 1448 | 873 | ±20.9 | 39.8 | $9.00 | |
| 19 | GPT-5.6 SolOpenAI | 1442 | 806 | ±21.9 | 73.5 | $20.00 | |
| 20 | GLM 5.2Z.ai | 1440 | 988 | ±19.9 | 61.1 | $4.40 | |
| 21 | GLM 5.3 FlashZ.ai | 1436 | 635 | ±24.2 | 57.7 | $0.50 | |
| 22 | DeepSeek V4.1 FlashDeepSeek | 1436 | 232 | ±38.9 | 62.5 | $0.84 | |
| 23 | MiMo-V2.5-ProXiaomi | 1435 | 1,407 | ±17.4 | 31.6 | $0.9152 | |
| 24 | Gemini 2.5 ProGoogle | 1434 | 2,497 | ±13.0 | 19.1 | $10.00 | |
| 25 | Claude Opus 4.8Anthropic | 1431 | 1,100 | ±19.3 | 65.0 | $25.00 | |
| 26 | Gemini 3.6 FlashGoogle | 1429 | 707 | ±22.6 | 62.5 | $3.75 | |
| 27 | GPT-6 AstraOpenAI | 1428 | 230 | ±39.3 | 85.1 | $50.00 | |
| 28 | Kimi K2.6Moonshot AI | 1427 | 734 | ±22.7 | 52.1 | $4.00 | |
| 29 | GPT-5.6 TerraOpenAI | 1427 | 854 | ±21.0 | 63.9 | $12.00 | |
| 30 | MiMo-V2.6-FlashXiaomi | 1425 | 250 | ±38.5 | 54.7 | $0.28 | |
| 31 | Claude Opus 4.5Anthropic | 1424 | 1,168 | ±18.2 | 44.0 | $25.00 | |
| 32 | GLM 5.1Z.ai | 1424 | 1,209 | ±18.4 | 49.2 | $4.40 | |
| 33 | GLM 5Z.ai | 1423 | 552 | ±25.0 | 41.7 | $3.20 | |
| 34 | GPT-5.6 LunaOpenAI | 1415 | 787 | ±21.5 | 58.0 | $1.20 | |
| 35 | Qwen3.7 PlusQwen | 1413 | 830 | ±21.3 | 33.2 | $1.28 | |
| 36 | Kimi K2.5Moonshot AI | 1412 | 1,355 | ±17.2 | 38.6 | $2.85 | |
| 37 | Claude Sonnet 5Anthropic | 1412 | 881 | ±20.8 | 57.7 | $10.00 | |
| 38 | Claude Sonnet 4.6Anthropic | 1410 | 1,148 | ±18.5 | 57.3 | $15.00 | |
| 39 | DeepSeek V4 Pro 0423DeepSeek | 1409 | 1,110 | ±18.9 | 38.7 | $2.976 | |
| 40 | Gemini 3.5 Flash LiteGoogle | 1407 | 714 | ±23.1 | 49.0 | $2.50 | |
| 41 | Grok 4.5xAI | 1406 | 729 | ±22.5 | 58.6 | $6.00 | |
| 42 | T InklingThinking Machines | 1403 | 769 | ±22.1 | 31.5 | $4.05 | |
| 43 | Claude Sonnet 4.5Anthropic | 1402 | 1,327 | ±17.0 | 39.5 | $15.00 | |
| 44 | GPT-5.1OpenAI | 1401 | 663 | ±23.2 | 40.4 | $10.00 | |
| 45 | GLM 4.6Z.ai | 1400 | 618 | ±24.7 | 8.8 | $2.20 | |
| 46 | GLM 4.7Z.ai | 1399 | 177 | ±43.1 | 55.1 | $2.20 | |
| 47 | GLM 5V TurboZ.ai | 1397 | 220 | ±38.0 | Not benchmarked | $4.00 | |
| 48 | Hy3Tencent | 1396 | 269 | ±35.8 | Not benchmarked | $0.528 | |
| 49 | o1OpenAI | 1396 | 396 | ±31.7 | Not benchmarked | $60.00 | |
| 50 | Qwen3 VL 235B A22B InstructQwen | 1394 | 325 | ±33.3 | Not benchmarked | $1.90 | |
| 51 | Gemini 3.1 Flash Lite PreviewGoogle | 1392 | 1,062 | ±19.1 | 22.7 | $1.50 | |
| 52 | Grok 4.6xAI | 1391 | 579 | ±25.6 | 65.7 | $6.00 | |
| 53 | Gemini 2.5 FlashGoogle | 1385 | 2,371 | ±13.2 | Not benchmarked | $2.50 | |
| 54 | Qwen3.5 397B A17BQwen | 1384 | 1,590 | ±16.2 | Not benchmarked | $2.34 | |
| 55 | Qwen3.8 27BQwen | 1384 | 407 | ±29.4 | 74.2 | $2.55 | |
| 56 | Mistral Medium 3.5Mistral AI | 1384 | 187 | ±43.5 | 25.5 | $7.50 | |
| 57 | Nemotron 3 UltraNVIDIA | 1384 | 211 | ±42.3 | 30.0 | $2.40 | |
| 58 | DeepSeek V4 Flash 0423DeepSeek | 1383 | 896 | ±20.9 | Not benchmarked | $0.28 | |
| 59 | Solar Pro 4Upstage | 1383 | 367 | ±30.8 | Not benchmarked | $0.60 | |
| 60 | Grok 4.7xAI | 1383 | 208 | ±41.6 | 66.7 | $6.00 | |
| 61 | Claude Opus 4.1Anthropic | 1381 | 1,102 | ±18.6 | Not benchmarked | $75.00 | |
| 62 | Mistral Medium 3.1Mistral AI | 1380 | 1,991 | ±14.1 | Not benchmarked | $2.00 | |
| 63 | GLM 4.5Z.ai | 1380 | 656 | ±23.9 | Not benchmarked | $2.20 | |
| 64 | Qwen3.6 PlusQwen | 1379 | 820 | ±21.0 | 44.1 | $1.95 | |
| 65 | Qwen3 235B A22B Thinking 2507Qwen | 1375 | 204 | ±41.0 | Not benchmarked | $3.50 | |
| 66 | GPT-5.2OpenAI | 1374 | 1,301 | ±17.3 | 55.9 | $14.00 | |
| 67 | MiniMax M3MiniMax | 1373 | 1,198 | ±18.6 | 39.2 | $1.20 | |
| 68 | DeepSeek V3.2 ExpDeepSeek | 1372 | 212 | ±39.4 | Not benchmarked | $0.41 | |
| 69 | GPT-5.4 MiniOpenAI | 1370 | 987 | ±19.9 | 35.8 | $4.50 | |
| 70 | DeepSeek V3.2DeepSeek | 1370 | 795 | ±21.3 | 30.0 | $0.88 | |
| 71 | o3OpenAI | 1370 | 1,201 | ±18.2 | Not benchmarked | $8.00 | |
| 72 | T Inkling SmallThinking Machines | 1367 | 600 | ±24.7 | 27.3 | $1.20 | |
| 73 | Qwen3 Next 80B A3B InstructQwen | 1363 | 523 | ±26.7 | Not benchmarked | $1.10 | |
| 74 | MiMo-V2.5Xiaomi | 1361 | 804 | ±21.6 | 50.0 | $0.336 | |
| 75 | GPT-5OpenAI | 1360 | 822 | ±21.3 | 38.8 | $10.00 | |
| 76 | R1 0528DeepSeek | 1360 | 323 | ±34.4 | Not benchmarked | $2.18 | |
| 77 | GPT-6 LunaOpenAI | 1359 | 967 | ±21.1 | 57.3 | $0.50 | |
| 78 | Qwen3.5-27BQwen | 1359 | 446 | ±28.8 | Not benchmarked | $1.56 | |
| 79 | Qwen3.5-35B-A3BQwen | 1355 | 482 | ±27.4 | Not benchmarked | $1.30 | |
| 80 | Grok 4.3xAI | 1354 | 1,125 | ±19.1 | 37.4 | $2.50 | |
| 81 | Qwen3.5-122B-A10BQwen | 1353 | 477 | ±27.4 | Not benchmarked | $2.08 | |
| 82 | Step 3.5 FlashStepFun | 1351 | 1,087 | ±18.8 | Not benchmarked | $0.30 | |
| 83 | Claude Haiku 4.5Anthropic | 1346 | 2,400 | ±13.5 | 35.6 | $5.00 | |
| 84 | Qwen3.5-FlashQwen | 1344 | 1,075 | ±19.1 | 41.1 | $0.26 | |
| 85 | GPT-4.1OpenAI | 1338 | 1,059 | ±18.8 | Not benchmarked | $8.00 | |
| 86 | GPT-6 SolOpenAI | 1336 | 228 | ±40.5 | 76.0 | $10.00 | |
| 87 | Qwen3 30B A3B Instruct 2507Qwen | 1330 | 595 | ±24.8 | Not benchmarked | $0.30 | |
| 88 | R1RetiredDeepSeek | 1329 | 215 | ±39.6 | Not benchmarked | $2.50 | |
| 89 | GPT-5.4 NanoOpenAI | 1321 | 1,033 | ±19.7 | 44.8 | $1.25 | |
| 90 | MiniMax M2.7MiniMax | 1314 | 1,545 | ±16.8 | 36.2 | $1.20 | |
| 91 | o3 MiniOpenAI | 1313 | 241 | ±40.8 | Not benchmarked | $4.40 | |
| 92 | Gemini 2.0 FlashRetiredGoogle | 1312 | 662 | ±23.4 | Not benchmarked | $0.40 | |
| 93 | o4 MiniOpenAI | 1312 | 986 | ±19.6 | Not benchmarked | $4.40 | |
| 94 | Qwen3 Next 80B A3B ThinkingQwen | 1310 | 356 | ±32.3 | Not benchmarked | $1.20 | |
| 95 | GLM 4.5 AirZ.ai | 1308 | 673 | ±23.1 | Not benchmarked | $1.10 | |
| 96 | GPT-5 MiniOpenAI | 1308 | 662 | ±24.2 | 34.2 | $2.00 | |
| 97 | Gemma 3 27BGoogle | 1308 | 889 | ±20.6 | Not benchmarked | $0.30 | |
| 98 | Trinity Large ThinkingArcee AI | 1306 | 532 | ±26.3 | Not benchmarked | $0.80 | |
| 99 | Qwen3 Coder 480B A35BQwen | 1305 | 612 | ±25.1 | Not benchmarked | $1.50 | |
| 100 | Mistral Medium 3Mistral AI | 1304 | 637 | ±23.8 | Not benchmarked | $2.00 | |
| 101 | Claude Opus 4RetiredAnthropic | 1303 | 1,031 | ±19.5 | Not benchmarked | $75.00 | |
| 102 | GPT-4.1 MiniOpenAI | 1298 | 807 | ±21.5 | Not benchmarked | $1.60 | |
| 103 | Gemini 1.5 ProRetiredGoogle | 1298 | 773 | ±23.6 | Not benchmarked | $5.00 | |
| 104 | Command ACohere | 1285 | 1,116 | ±18.3 | Not benchmarked | $10.00 | |
| 105 | GLM 4.7 FlashZ.ai | 1283 | 239 | ±40.4 | Not benchmarked | $0.40 | |
| 106 | gpt-oss-120bOpenAI | 1282 | 681 | ±23.8 | Not benchmarked | $0.50 | |
| 107 | Nova 2 LiteAmazon | 1281 | 161 | ±48.1 | Not benchmarked | $2.50 | |
| 108 | GPT-5 NanoOpenAI | 1269 | 225 | ±40.6 | Not benchmarked | $0.40 | |
| 109 | Nemotron 3 Nano 30B A3BNVIDIA | 1269 | 241 | ±40.0 | Not benchmarked | $0.22 | |
| 110 | MiniMax M1MiniMax | 1266 | 672 | ±23.9 | Not benchmarked | $2.20 | |
| 111 | Qwen3 30B A3BQwen | 1261 | 475 | ±27.0 | Not benchmarked | $0.50 | |
| 112 | Claude 3.7 SonnetRetiredAnthropic | 1239 | 702 | ±23.6 | Not benchmarked | $15.00 | |
| 113 | gpt-oss-20bOpenAI | 1236 | 228 | ±40.9 | Not benchmarked | $0.15 | |
| 114 | GPT-4o (2024-05-13)OpenAI | 1234 | 3,517 | ±17.4 | Not benchmarked | $15.00 | |
| 115 | MiniMax M2.5MiniMax | 1233 | 729 | ±24.1 | 37.2 | $1.20 | |
| 116 | o1-miniRetiredOpenAI | 1223 | 716 | ±23.9 | Not benchmarked | $4.40 | |
| 117 | Gemini 1.5 FlashRetiredGoogle | 1220 | 502 | ±28.5 | Not benchmarked | $0.30 | |
| 118 | GPT-4o (2024-08-06)OpenAI | 1218 | 712 | ±24.8 | Not benchmarked | $10.00 | |
| 119 | Llama 4 ScoutMeta Llama | 1207 | 649 | ±25.2 | Not benchmarked | $0.50 | |
| 120 | Mistral Small 3.1 24BMistral AI | 1207 | 713 | ±23.9 | Not benchmarked | $0.555 | |
| 121 | Llama 4 MaverickMeta Llama | 1203 | 796 | ±21.9 | Not benchmarked | $1.00 | |
| 122 | Nova Pro 1.0Amazon | 1203 | 326 | ±32.6 | Not benchmarked | $3.20 | |
| 123 | Mistral LargeMistral AI | 1202 | 408 | ±29.9 | Not benchmarked | $6.00 | |
| 124 | Claude 3.5 SonnetRetiredAnthropic | 1200 | 1,715 | ±19.5 | Not benchmarked | $15.00 | |
| 125 | GPT-4o-mini (2024-07-18)OpenAI | 1195 | 1,087 | ±20.5 | Not benchmarked | $0.60 | |
| 126 | Mistral Small 3Mistral AI | 1188 | 231 | ±39.6 | Not benchmarked | $0.08 | |
| 127 | Qwen2.5 72B InstructQwen | 1188 | 527 | ±27.8 | Not benchmarked | $0.40 | |
| 128 | Claude 3 OpusRetiredAnthropic | 1187 | 3,906 | ±17.1 | Not benchmarked | $75.00 | |
| 129 | GPT-4 TurboOpenAI | 1187 | 2,668 | ±18.6 | Not benchmarked | $30.00 | |
| 130 | Gemma 2 27BGoogle | 1174 | 1,440 | ±19.3 | Not benchmarked | $0.65 | |
| 131 | Claude 3.5 HaikuRetiredAnthropic | 1173 | 1,213 | ±18.6 | Not benchmarked | $4.00 | |
| 132 | Command R (08-2024)Cohere | 1162 | 152 | ±50.9 | Not benchmarked | $0.60 | |
| 133 | Nova Lite 1.0Amazon | 1153 | 292 | ±35.7 | Not benchmarked | $0.24 | |
| 134 | Phi 4Microsoft | 1151 | 319 | ±33.1 | Not benchmarked | $0.14 | |
| 135 | Nova Micro 1.0Amazon | 1150 | 267 | ±37.8 | Not benchmarked | $0.14 | |
| 136 | Llama 3.3 70B InstructMeta Llama | 1142 | 856 | ±21.9 | Not benchmarked | $0.72 | |
| 137 | Llama 3.1 70B InstructMeta Llama | 1140 | 873 | ±23.0 | Not benchmarked | $0.56 | |
| 138 | Claude 3 SonnetRetiredAnthropic | 1128 | 2,059 | ±19.4 | Not benchmarked | $15.00 | |
| 139 | Claude 3 HaikuRetiredAnthropic | 1108 | 2,364 | ±18.8 | Not benchmarked | $1.25 | |
| 140 | GPT-4OpenAI | 1087 | 429 | ±32.1 | Not benchmarked | $60.00 | |
| 141 | Llama 3.1 8B InstructMeta Llama | 1053 | 802 | ±24.1 | 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 22, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.