LMArena Text: Spanish leaderboard
- Claude Opus 5 leads
- 1523
- models measured
- 139
- lead over #2
- +3 Elo
- median
- 1422
Leader on LMArena Text: Spanish · read October 22
On October 22, 2026, 151 of our catalog's models had a LMArena Text: Spanish result, and Anthropic Claude Opus 5 led with 1523, ahead of Anthropic Claude Fable 5.1 at 1520.
Prompts written in Spanish. 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: Spanish 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 | 1523 | 993 | ±19.7 | 83.0 | $25.00 | |
| 2 | Claude Fable 5.1Anthropic | 1520 | 507 | ±25.5 | 84.2 | $50.00 | |
| 3 | Claude Opus 4.6Anthropic | 1510 | 2,889 | ±12.3 | 57.2 | $25.00 | |
| 4 | Claude Opus 5.5Anthropic | 1508 | 335 | ±31.7 | 88.8 | $20.00 | |
| 5 | Gemini 4 ArgonGoogle | 1503 | 224 | ±38.2 | 81.0 | $10.00 | |
| 6 | Claude Fable 5Anthropic | 1498 | 1,361 | ±16.8 | 79.3 | $50.00 | |
| 7 | Muse Spark 1.3Meta | 1496 | 617 | ±24.3 | 57.4 | $4.25 | |
| 8 | Gemini 3.7 FlashGoogle | 1495 | 928 | ±20.3 | 58.0 | $3.75 | |
| 9 | Claude Opus 4.7Anthropic | 1494 | 2,055 | ±14.3 | 67.5 | $25.00 | |
| 10 | Qwen3.8 Max (0902)Qwen | 1489 | 930 | ±20.0 | 62.4 | $6.00 | |
| 11 | Gemini 3.8 FlashGoogle | 1487 | 1,425 | ±16.4 | 61.6 | $3.75 | |
| 12 | Muse Spark 1.2Meta | 1484 | 308 | ±32.4 | 55.7 | $4.25 | |
| 13 | Gemini 3.1 Pro PreviewGoogle | 1481 | 4,141 | ±10.6 | 32.3 | $12.00 | |
| 14 | Gemini 3.5 FlashGoogle | 1480 | 1,427 | ±16.6 | 39.8 | $9.00 | |
| 15 | GLM 5.2Z.ai | 1478 | 1,548 | ±15.8 | 61.1 | $4.40 | |
| 16 | Gemini 3.6 FlashGoogle | 1476 | 947 | ±20.2 | 62.5 | $3.75 | |
| 17 | GLM 5.3 FlashZ.ai | 1475 | 1,360 | ±16.8 | 57.7 | $0.50 | |
| 18 | Gemini 2.5 ProGoogle | 1474 | 3,453 | ±11.0 | 19.1 | $10.00 | |
| 19 | Kimi K3Moonshot AI | 1472 | 1,059 | ±18.4 | 62.3 | $15.00 | |
| 20 | MiMo-V2.5-ProXiaomi | 1469 | 2,383 | ±13.3 | 31.6 | $0.9152 | |
| 21 | GLM 5.1Z.ai | 1468 | 2,112 | ±13.7 | 49.2 | $4.40 | |
| 22 | GPT-5.5OpenAI | 1468 | 1,945 | ±14.3 | 71.3 | $30.00 | |
| 23 | Claude Opus 4.8Anthropic | 1465 | 1,860 | ±14.7 | 65.0 | $25.00 | |
| 24 | Claude Sonnet 4.6Anthropic | 1464 | 2,168 | ±13.8 | 57.3 | $15.00 | |
| 25 | Kimi K2.6Moonshot AI | 1464 | 1,371 | ±16.6 | 52.1 | $4.00 | |
| 26 | Muse Spark 1.1Meta | 1464 | 1,406 | ±16.5 | 70.3 | $4.25 | |
| 27 | GLM 5.3Z.ai | 1462 | 772 | ±20.9 | 70.6 | $4.40 | |
| 28 | DeepSeek V4.1 FlashDeepSeek | 1459 | 465 | ±26.8 | 62.5 | $0.84 | |
| 29 | Claude Sonnet 4.5Anthropic | 1458 | 2,500 | ±12.4 | 39.5 | $15.00 | |
| 30 | Claude Opus 4.5Anthropic | 1457 | 1,147 | ±17.7 | 44.0 | $25.00 | |
| 31 | Qwen3.7 PlusQwen | 1457 | 1,243 | ±17.6 | 33.2 | $1.28 | |
| 32 | Hy3Tencent | 1457 | 458 | ±27.2 | Not benchmarked | $0.528 | |
| 33 | DeepSeek V4 Pro 0423DeepSeek | 1457 | 1,732 | ±15.1 | 38.7 | $2.976 | |
| 34 | Mistral Large 4Mistral AI | 1455 | 320 | ±32.9 | 56.4 | $4.18 | |
| 35 | Nemotron 3 UltraNVIDIA | 1455 | 453 | ±28.3 | 30.0 | $2.40 | |
| 36 | GLM 4.5Z.ai | 1454 | 678 | ±23.3 | Not benchmarked | $2.20 | |
| 37 | GLM 5Z.ai | 1454 | 1,131 | ±17.9 | 41.7 | $3.20 | |
| 38 | MiMo-V2.6-FlashXiaomi | 1453 | 471 | ±27.0 | 54.7 | $0.28 | |
| 39 | GPT-5.4OpenAI | 1452 | 2,036 | ±14.2 | 64.4 | $15.00 | |
| 40 | GLM 5V TurboZ.ai | 1452 | 299 | ±35.5 | Not benchmarked | $4.00 | |
| 41 | Grok 4.5xAI | 1451 | 1,080 | ±18.5 | 58.6 | $6.00 | |
| 42 | Claude Sonnet 5.5Anthropic | 1451 | 286 | ±33.4 | 87.3 | $10.00 | |
| 43 | GPT-5.6 TerraOpenAI | 1451 | 1,378 | ±16.4 | 63.9 | $12.00 | |
| 44 | Kimi K2.5Moonshot AI | 1450 | 2,535 | ±12.7 | 38.6 | $2.85 | |
| 45 | T InklingThinking Machines | 1449 | 1,338 | ±16.9 | 31.5 | $4.05 | |
| 46 | Qwen3.8 27BQwen | 1449 | 807 | ±21.2 | 74.2 | $2.55 | |
| 47 | Claude Opus 4.1Anthropic | 1448 | 2,302 | ±13.3 | Not benchmarked | $75.00 | |
| 48 | GPT-5.6 LunaOpenAI | 1446 | 1,189 | ±17.8 | 58.0 | $1.20 | |
| 49 | GPT-5.6 SolOpenAI | 1443 | 1,229 | ±17.3 | 73.5 | $20.00 | |
| 50 | Gemma 4 31BGoogle | 1442 | 251 | ±35.6 | 39.0 | $0.40 | |
| 51 | DeepSeek V3.2 ExpDeepSeek | 1441 | 355 | ±30.9 | Not benchmarked | $0.41 | |
| 52 | Gemini 3.5 Flash LiteGoogle | 1441 | 965 | ±19.6 | 49.0 | $2.50 | |
| 53 | Qwen3.5 397B A17BQwen | 1441 | 2,989 | ±11.8 | Not benchmarked | $2.34 | |
| 54 | Step 5 PreviewStepFun | 1441 | 318 | ±31.9 | 61.1 | $2.70 | |
| 55 | GPT-6.1 SolOpenAI | 1439 | 257 | ±36.3 | 82.4 | $10.00 | |
| 56 | Claude Sonnet 5Anthropic | 1438 | 1,455 | ±16.3 | 57.7 | $10.00 | |
| 57 | GLM 4.6Z.ai | 1436 | 1,014 | ±19.0 | 8.8 | $2.20 | |
| 58 | DeepSeek V4 Flash 0423DeepSeek | 1436 | 1,580 | ±15.9 | Not benchmarked | $0.28 | |
| 59 | DeepSeek V4 Pro 0813DeepSeek | 1435 | 299 | ±32.3 | 56.1 | $1.98 | |
| 60 | Qwen3 Next 80B A3B InstructQwen | 1435 | 688 | ±23.0 | Not benchmarked | $1.10 | |
| 61 | GLM 4.7Z.ai | 1434 | 345 | ±30.8 | 55.1 | $2.20 | |
| 62 | GPT-5.1OpenAI | 1433 | 1,208 | ±17.3 | 40.4 | $10.00 | |
| 63 | MiniMax M3MiniMax | 1433 | 2,040 | ±14.3 | 39.2 | $1.20 | |
| 64 | Mistral Medium 3.1Mistral AI | 1433 | 2,867 | ±11.8 | Not benchmarked | $2.00 | |
| 65 | Qwen3.6 PlusQwen | 1432 | 1,578 | ±15.8 | 44.1 | $1.95 | |
| 66 | Qwen3 VL 235B A22B InstructQwen | 1428 | 460 | ±28.1 | Not benchmarked | $1.90 | |
| 67 | Qwen3.5-122B-A10BQwen | 1425 | 968 | ±18.9 | Not benchmarked | $2.08 | |
| 68 | DeepSeek V3.2DeepSeek | 1424 | 1,531 | ±15.7 | 30.0 | $0.88 | |
| 69 | T Inkling SmallThinking Machines | 1422 | 1,022 | ±19.1 | 27.3 | $1.20 | |
| 70 | Gemini 2.5 FlashGoogle | 1422 | 3,460 | ±10.9 | Not benchmarked | $2.50 | |
| 71 | GPT-6 AstraOpenAI | 1422 | 416 | ±28.8 | 85.1 | $50.00 | |
| 72 | Claude Haiku 4.5Anthropic | 1421 | 4,444 | ±10.2 | 35.6 | $5.00 | |
| 73 | Gemini 3.1 Flash Lite PreviewGoogle | 1421 | 2,020 | ±13.9 | 22.7 | $1.50 | |
| 74 | Step 3.5 FlashStepFun | 1419 | 2,014 | ±13.8 | Not benchmarked | $0.30 | |
| 75 | Nemotron 3 SuperNVIDIA | 1418 | 352 | ±30.8 | Not benchmarked | $0.45 | |
| 76 | Gemma 4 26B A4BGoogle | 1415 | 201 | ±41.7 | Not benchmarked | $0.60 | |
| 77 | Grok 4.6xAI | 1414 | 862 | ±20.3 | 65.7 | $6.00 | |
| 78 | MiMo-V2.5Xiaomi | 1414 | 1,442 | ±16.6 | 50.0 | $0.336 | |
| 79 | GPT-5.2OpenAI | 1413 | 1,536 | ±15.6 | 55.9 | $14.00 | |
| 80 | R1 0528DeepSeek | 1411 | 261 | ±35.5 | Not benchmarked | $2.18 | |
| 81 | Mistral Medium 3.5Mistral AI | 1411 | 389 | ±31.0 | 25.5 | $7.50 | |
| 82 | Solar Pro 4Upstage | 1409 | 715 | ±22.8 | Not benchmarked | $0.60 | |
| 83 | Qwen3.5-27BQwen | 1407 | 940 | ±19.4 | Not benchmarked | $1.56 | |
| 84 | Grok 4.7xAI | 1406 | 477 | ±26.9 | 66.7 | $6.00 | |
| 85 | GPT-5.4 MiniOpenAI | 1405 | 1,915 | ±14.7 | 35.8 | $4.50 | |
| 86 | Qwen3 30B A3B Instruct 2507Qwen | 1404 | 716 | ±22.4 | Not benchmarked | $0.30 | |
| 87 | MiniMax M2.7MiniMax | 1402 | 2,781 | ±12.5 | 36.2 | $1.20 | |
| 88 | Qwen3.5-FlashQwen | 1400 | 2,055 | ±14.1 | 41.1 | $0.26 | |
| 89 | Grok 4.3xAI | 1399 | 2,009 | ±14.5 | 37.4 | $2.50 | |
| 90 | GPT-5OpenAI | 1399 | 901 | ±20.3 | 38.8 | $10.00 | |
| 91 | Hy3 previewTencent | 1399 | 210 | ±39.9 | Not benchmarked | $0.60 | |
| 92 | o3OpenAI | 1395 | 1,221 | ±17.9 | Not benchmarked | $8.00 | |
| 93 | GPT-6 LunaOpenAI | 1391 | 2,391 | ±13.7 | 57.3 | $0.50 | |
| 94 | Qwen3.5-35B-A3BQwen | 1390 | 1,007 | ±19.2 | Not benchmarked | $1.30 | |
| 95 | gpt-oss-120bOpenAI | 1389 | 855 | ±21.2 | Not benchmarked | $0.50 | |
| 96 | Qwen3 235B A22B Thinking 2507Qwen | 1389 | 185 | ±41.0 | Not benchmarked | $3.50 | |
| 97 | GPT-6 SolOpenAI | 1387 | 552 | ±25.9 | 76.0 | $10.00 | |
| 98 | GLM 4.5 AirZ.ai | 1385 | 824 | ±21.0 | Not benchmarked | $1.10 | |
| 99 | R1RetiredDeepSeek | 1379 | 114 | ±53.0 | Not benchmarked | $2.50 | |
| 100 | GPT-4.1OpenAI | 1376 | 1,132 | ±18.1 | Not benchmarked | $8.00 | |
| 101 | Nemotron 3 Nano 30B A3BNVIDIA | 1376 | 362 | ±31.1 | Not benchmarked | $0.22 | |
| 102 | Nova 2 LiteAmazon | 1375 | 230 | ±39.4 | Not benchmarked | $2.50 | |
| 103 | GPT-5.4 NanoOpenAI | 1371 | 1,920 | ±14.9 | 44.8 | $1.25 | |
| 104 | Mistral Medium 3Mistral AI | 1364 | 567 | ±25.1 | Not benchmarked | $2.00 | |
| 105 | Gemini 2.0 FlashRetiredGoogle | 1363 | 553 | ±25.0 | Not benchmarked | $0.40 | |
| 106 | GPT-5 NanoOpenAI | 1360 | 237 | ±39.2 | Not benchmarked | $0.40 | |
| 107 | Qwen3 Coder 480B A35BQwen | 1360 | 686 | ±23.3 | Not benchmarked | $1.50 | |
| 108 | Qwen3 Next 80B A3B ThinkingQwen | 1358 | 456 | ±28.1 | Not benchmarked | $1.20 | |
| 109 | Claude Opus 4RetiredAnthropic | 1357 | 1,001 | ±19.6 | Not benchmarked | $75.00 | |
| 110 | GPT-5 MiniOpenAI | 1354 | 856 | ±20.9 | 34.2 | $2.00 | |
| 111 | MiniMax M1MiniMax | 1353 | 835 | ±20.9 | Not benchmarked | $2.20 | |
| 112 | MiniMax M2.5MiniMax | 1353 | 1,478 | ±16.5 | 37.2 | $1.20 | |
| 113 | GLM 4.7 FlashZ.ai | 1350 | 439 | ±28.4 | Not benchmarked | $0.40 | |
| 114 | Gemma 3 27BGoogle | 1349 | 893 | ±20.0 | Not benchmarked | $0.30 | |
| 115 | Command ACohere | 1347 | 1,179 | ±17.5 | Not benchmarked | $10.00 | |
| 116 | o4 MiniOpenAI | 1346 | 996 | ±19.4 | Not benchmarked | $4.40 | |
| 117 | o1OpenAI | 1345 | 142 | ±48.2 | Not benchmarked | $60.00 | |
| 118 | GLM 4.5VZ.ai | 1336 | 186 | ±44.9 | Not benchmarked | $1.80 | |
| 119 | Trinity Large ThinkingArcee AI | 1327 | 948 | ±20.6 | Not benchmarked | $0.80 | |
| 120 | MiniMax M2MiniMax | 1326 | 206 | ±42.8 | Not benchmarked | $1.02 | |
| 121 | o3 MiniOpenAI | 1321 | 126 | ±50.1 | Not benchmarked | $4.40 | |
| 122 | GPT-4.1 MiniOpenAI | 1319 | 805 | ±21.0 | Not benchmarked | $1.60 | |
| 123 | Qwen3 30B A3BQwen | 1317 | 481 | ±27.4 | Not benchmarked | $0.50 | |
| 124 | Gemini 1.5 ProRetiredGoogle | 1310 | 405 | ±29.3 | Not benchmarked | $5.00 | |
| 125 | o1-miniRetiredOpenAI | 1303 | 425 | ±27.9 | Not benchmarked | $4.40 | |
| 126 | Llama 4 MaverickMeta Llama | 1293 | 820 | ±21.4 | Not benchmarked | $1.00 | |
| 127 | GPT-4o (2024-05-13)OpenAI | 1292 | 1,712 | ±20.2 | Not benchmarked | $15.00 | |
| 128 | Claude 3.7 SonnetRetiredAnthropic | 1290 | 623 | ±24.6 | Not benchmarked | $15.00 | |
| 129 | Claude 3.5 SonnetRetiredAnthropic | 1289 | 1,123 | ±22.4 | Not benchmarked | $15.00 | |
| 130 | Mistral Small 3.1 24BMistral AI | 1284 | 809 | ±21.8 | Not benchmarked | $0.555 | |
| 131 | Llama 4 ScoutMeta Llama | 1278 | 711 | ±23.8 | Not benchmarked | $0.50 | |
| 132 | GPT-4o-mini (2024-07-18)OpenAI | 1276 | 741 | ±23.8 | Not benchmarked | $0.60 | |
| 133 | Llama 3.3 70B InstructMeta Llama | 1269 | 636 | ±24.0 | Not benchmarked | $0.72 | |
| 134 | GPT-4o (2024-08-06)OpenAI | 1269 | 546 | ±27.1 | Not benchmarked | $10.00 | |
| 135 | Mistral LargeMistral AI | 1268 | 153 | ±45.4 | Not benchmarked | $6.00 | |
| 136 | gpt-oss-20bOpenAI | 1267 | 284 | ±38.1 | Not benchmarked | $0.15 | |
| 137 | Claude 3.5 HaikuRetiredAnthropic | 1261 | 963 | ±20.0 | Not benchmarked | $4.00 | |
| 138 | GPT-4 TurboOpenAI | 1260 | 1,430 | ±21.2 | Not benchmarked | $30.00 | |
| 139 | Qwen2.5 72B InstructQwen | 1255 | 347 | ±30.9 | Not benchmarked | $0.40 | |
| 140 | Llama 3.1 70B InstructMeta Llama | 1253 | 606 | ±25.2 | Not benchmarked | $0.56 | |
| 141 | Claude 3 OpusRetiredAnthropic | 1246 | 2,569 | ±19.1 | Not benchmarked | $75.00 | |
| 142 | Gemini 1.5 FlashRetiredGoogle | 1243 | 283 | ±33.7 | Not benchmarked | $0.30 | |
| 143 | Phi 4Microsoft | 1234 | 127 | ±51.4 | Not benchmarked | $0.14 | |
| 144 | Gemma 2 27BGoogle | 1228 | 845 | ±22.8 | Not benchmarked | $0.65 | |
| 145 | Nova Lite 1.0Amazon | 1226 | 107 | ±53.1 | Not benchmarked | $0.24 | |
| 146 | Nova Micro 1.0Amazon | 1225 | 144 | ±45.6 | Not benchmarked | $0.14 | |
| 147 | Claude 3 SonnetRetiredAnthropic | 1204 | 1,448 | ±21.9 | Not benchmarked | $15.00 | |
| 148 | GPT-4OpenAI | 1197 | 647 | ±27.7 | Not benchmarked | $60.00 | |
| 149 | Nova Pro 1.0Amazon | 1182 | 150 | ±49.8 | Not benchmarked | $3.20 | |
| 150 | Llama 3.1 8B InstructMeta Llama | 1169 | 565 | ±27.3 | Not benchmarked | $0.08 | |
| 151 | Claude 3 HaikuRetiredAnthropic | 1165 | 1,704 | ±20.6 | Not benchmarked | $1.25 |
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.