LMArena Text: legal and government leaderboard
- Gemini 4 Argon leads
- 1548
- models measured
- 155
- lead over #2
- +18 Elo
- median
- 1425
Leader on LMArena Text: legal and government · read October 21
On October 21, 2026, 167 of our catalog's models had a LMArena Text: legal and government result, and Google Gemini 4 Argon led with 1548, ahead of Anthropic Claude Opus 5 at 1530.
Prompts from legal and government work. 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: legal and government 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 | Gemini 4 ArgonGoogle | 1548 | 399 | ±30.1 | 81.0 | $10.00 | |
| 2 | Claude Opus 5Anthropic | 1530 | 2,901 | ±11.7 | 83.0 | $25.00 | |
| 3 | Claude Opus 5.5Anthropic | 1518 | 821 | ±21.6 | 88.8 | $20.00 | |
| 4 | Muse Spark 1.2Meta | 1517 | 757 | ±21.7 | 55.7 | $4.25 | |
| 5 | Claude Fable 5.1Anthropic | 1514 | 1,290 | ±17.0 | 84.2 | $50.00 | |
| 6 | Claude Opus 4.6Anthropic | 1512 | 6,744 | ±8.1 | 57.2 | $25.00 | |
| 7 | Muse WonderMeta | 1505 | 441 | ±28.7 | 71.9 | $20.00 | |
| 8 | Claude Opus 4.7Anthropic | 1504 | 5,261 | ±9.0 | 67.5 | $25.00 | |
| 9 | Kimi K3Moonshot AI | 1503 | 2,682 | ±11.9 | 62.3 | $15.00 | |
| 10 | Muse Spark 1.3Meta | 1499 | 1,615 | ±15.1 | 57.4 | $4.25 | |
| 11 | Claude Fable 5Anthropic | 1498 | 3,627 | ±10.5 | 79.3 | $50.00 | |
| 12 | Gemini 3.8 FlashGoogle | 1497 | 3,650 | ±10.5 | 61.6 | $3.75 | |
| 13 | Gemini 3.7 FlashGoogle | 1496 | 2,461 | ±12.5 | 58.0 | $3.75 | |
| 14 | Qwen3.8 Max (0902)Qwen | 1492 | 2,183 | ±13.1 | 62.4 | $6.00 | |
| 15 | GPT-5.5OpenAI | 1491 | 5,610 | ±8.7 | 71.3 | $30.00 | |
| 16 | Gemini 3.5 FlashGoogle | 1488 | 3,844 | ±10.2 | 39.8 | $9.00 | |
| 17 | Gemini 3.1 Pro PreviewGoogle | 1487 | 10,646 | ±6.7 | 32.3 | $12.00 | |
| 18 | GLM 5.2Z.ai | 1486 | 4,161 | ±9.8 | 61.1 | $4.40 | |
| 19 | GPT-5.4OpenAI | 1486 | 5,232 | ±9.0 | 64.4 | $15.00 | |
| 20 | MiMo-V2.6-ProXiaomi | 1485 | 345 | ±31.8 | 65.4 | $0.87 | |
| 21 | Claude Sonnet 5.5Anthropic | 1483 | 698 | ±22.6 | 87.3 | $10.00 | |
| 22 | Muse Spark 1.1Meta | 1481 | 3,853 | ±10.2 | 70.3 | $4.25 | |
| 23 | Gemini 2.5 ProGoogle | 1480 | 8,929 | ±6.7 | 19.1 | $10.00 | |
| 24 | Gemini 3.6 FlashGoogle | 1479 | 3,017 | ±11.2 | 62.5 | $3.75 | |
| 25 | Claude Opus 4.8Anthropic | 1477 | 5,523 | ±8.7 | 65.0 | $25.00 | |
| 26 | GPT-5.6 SolOpenAI | 1476 | 3,544 | ±10.5 | 73.5 | $20.00 | |
| 27 | MiMo-V2.5-ProXiaomi | 1474 | 6,257 | ±8.2 | 31.6 | $0.9152 | |
| 28 | GLM 5.3Z.ai | 1473 | 2,096 | ±13.4 | 70.6 | $4.40 | |
| 29 | GLM 5.3 FlashZ.ai | 1473 | 2,957 | ±11.5 | 57.7 | $0.50 | |
| 30 | GPT-6.1 SolOpenAI | 1471 | 636 | ±23.6 | 82.4 | $10.00 | |
| 31 | Claude Sonnet 4.6Anthropic | 1470 | 5,709 | ±8.6 | 57.3 | $15.00 | |
| 32 | GLM 5.1Z.ai | 1465 | 5,083 | ±8.8 | 49.2 | $4.40 | |
| 33 | DeepSeek V4 Pro 0423DeepSeek | 1464 | 4,496 | ±9.4 | 38.7 | $2.716 | |
| 34 | Grok 4.5xAI | 1463 | 3,340 | ±10.9 | 58.6 | $6.00 | |
| 35 | GPT-6 AstraOpenAI | 1462 | 1,142 | ±17.7 | 85.1 | $50.00 | |
| 36 | Kimi K2.6Moonshot AI | 1462 | 3,023 | ±11.0 | 52.1 | $4.00 | |
| 37 | GPT-5.6 TerraOpenAI | 1461 | 3,831 | ±10.2 | 63.9 | $12.00 | |
| 38 | DeepSeek V4 Pro 0813DeepSeek | 1459 | 825 | ±20.3 | 56.1 | $1.98 | |
| 39 | Claude Opus 4.5Anthropic | 1459 | 5,430 | ±8.5 | 44.0 | $25.00 | |
| 40 | GPT-5.1OpenAI | 1459 | 3,067 | ±10.8 | 40.4 | $10.00 | |
| 41 | DeepSeek V3.1 TerminusDeepSeek | 1457 | 258 | ±35.5 | Not benchmarked | $1.0142 | |
| 42 | Gemma 4 26B A4BGoogle | 1457 | 416 | ±27.2 | Not benchmarked | $0.60 | |
| 43 | Claude Sonnet 5Anthropic | 1454 | 4,373 | ±9.6 | 57.7 | $10.00 | |
| 44 | GLM 5Z.ai | 1453 | 2,269 | ±12.6 | 41.7 | $3.20 | |
| 45 | Nemotron 3 UltraNVIDIA | 1452 | 969 | ±19.3 | 30.0 | $2.40 | |
| 46 | Qwen3.7 PlusQwen | 1451 | 3,560 | ±10.3 | 33.2 | $1.28 | |
| 47 | Claude Sonnet 4.5Anthropic | 1451 | 5,992 | ±8.0 | 39.5 | $15.00 | |
| 48 | GLM 4.6Z.ai | 1448 | 2,449 | ±11.8 | 8.8 | $2.20 | |
| 49 | T InklingThinking Machines | 1448 | 3,522 | ±10.5 | 31.5 | $4.05 | |
| 50 | Qwen3.8 27BQwen | 1447 | 1,945 | ±13.8 | 74.2 | $2.55 | |
| 51 | GPT-5.6 LunaOpenAI | 1446 | 3,481 | ±10.6 | 58.0 | $1.20 | |
| 52 | GLM 4.7Z.ai | 1444 | 969 | ±18.5 | 55.1 | $2.20 | |
| 53 | DeepSeek V4.1 FlashDeepSeek | 1443 | 1,145 | ±17.4 | 62.5 | $0.78 | |
| 54 | MiMo-V2.6-FlashXiaomi | 1441 | 1,061 | ±18.4 | 54.7 | $0.28 | |
| 55 | Kimi K2.5Moonshot AI | 1441 | 5,645 | ±8.3 | 38.6 | $2.85 | |
| 56 | Qwen3.5 397B A17BQwen | 1441 | 7,335 | ±7.5 | Not benchmarked | $2.34 | |
| 57 | Mistral Medium 3.1Mistral AI | 1439 | 6,698 | ±7.4 | Not benchmarked | $2.00 | |
| 58 | DeepSeek V4 Flash 0423DeepSeek | 1439 | 4,146 | ±9.7 | Not benchmarked | $0.28 | |
| 59 | Qwen3 VL 235B A22B InstructQwen | 1438 | 736 | ±22.4 | Not benchmarked | $1.90 | |
| 60 | Gemma 4 31BGoogle | 1438 | 451 | ±26.1 | 39.0 | $0.40 | |
| 61 | GLM 4.5Z.ai | 1438 | 1,476 | ±15.7 | Not benchmarked | $2.20 | |
| 62 | Step 5 PreviewStepFun | 1437 | 748 | ±21.3 | 61.1 | $2.70 | |
| 63 | Mistral Medium 3.5Mistral AI | 1437 | 891 | ±20.2 | 25.5 | $7.50 | |
| 64 | MiniMax M3MiniMax | 1437 | 5,344 | ±8.8 | 39.2 | $1.20 | |
| 65 | Qwen3.6 PlusQwen | 1437 | 3,620 | ±10.3 | 44.1 | $1.95 | |
| 66 | GLM 5V TurboZ.ai | 1436 | 830 | ±20.4 | Not benchmarked | $4.00 | |
| 67 | R Beam 1.0 501B A23BReflection | 1433 | 346 | ±31.3 | Not benchmarked | $1.10 | |
| 68 | Gemini 3.5 Flash LiteGoogle | 1431 | 3,053 | ±11.3 | 49.0 | $2.50 | |
| 69 | Qwen3 Next 80B A3B InstructQwen | 1431 | 1,415 | ±15.6 | Not benchmarked | $1.10 | |
| 70 | GPT-5OpenAI | 1431 | 1,901 | ±13.7 | 38.8 | $10.00 | |
| 71 | Hy3 previewTencent | 1430 | 555 | ±25.6 | Not benchmarked | $0.60 | |
| 72 | Gemini 2.5 FlashGoogle | 1430 | 8,913 | ±6.5 | Not benchmarked | $2.50 | |
| 73 | Grok 4.6xAI | 1429 | 2,643 | ±11.9 | 65.7 | $6.00 | |
| 74 | o3OpenAI | 1428 | 3,744 | ±10.2 | Not benchmarked | $8.00 | |
| 75 | GPT-5.2OpenAI | 1427 | 3,768 | ±9.9 | 55.9 | $14.00 | |
| 76 | DeepSeek V3.2DeepSeek | 1425 | 3,487 | ±10.2 | 30.0 | $0.88 | |
| 77 | Mistral Large 4Mistral AI | 1425 | 770 | ±21.3 | 56.4 | $4.18 | |
| 78 | MiMo-V2.5Xiaomi | 1425 | 3,770 | ±10.3 | 50.0 | $0.336 | |
| 79 | Grok 4.7xAI | 1423 | 1,162 | ±17.5 | 66.7 | $6.00 | |
| 80 | Hy3Tencent | 1422 | 978 | ±19.2 | Not benchmarked | $0.33 | |
| 81 | Qwen3.5-122B-A10BQwen | 1422 | 2,344 | ±12.4 | Not benchmarked | $2.08 | |
| 82 | Claude Opus 4.1Anthropic | 1421 | 3,215 | ±10.5 | Not benchmarked | $75.00 | |
| 83 | DeepSeek V3.2 ExpDeepSeek | 1419 | 502 | ±25.0 | Not benchmarked | $0.41 | |
| 84 | R1 0528DeepSeek | 1419 | 1,369 | ±16.2 | Not benchmarked | $2.18 | |
| 85 | Gemini 3.1 Flash Lite PreviewGoogle | 1418 | 5,018 | ±8.9 | 22.7 | $1.50 | |
| 86 | Qwen3 235B A22B Thinking 2507Qwen | 1415 | 532 | ±24.5 | Not benchmarked | $3.50 | |
| 87 | T Inkling SmallThinking Machines | 1415 | 2,956 | ±11.5 | 27.3 | $1.20 | |
| 88 | GPT-5.4 MiniOpenAI | 1414 | 5,075 | ±9.1 | 35.8 | $4.50 | |
| 89 | Step 3.5 FlashStepFun | 1408 | 4,661 | ±9.1 | Not benchmarked | $0.30 | |
| 90 | Qwen3.5-27BQwen | 1405 | 2,185 | ±12.8 | Not benchmarked | $1.56 | |
| 91 | MiniMax M2.7MiniMax | 1403 | 6,648 | ±8.1 | 36.2 | $1.20 | |
| 92 | Claude Haiku 4.5Anthropic | 1400 | 11,370 | ±6.2 | 35.6 | $5.00 | |
| 93 | GPT-6 LunaOpenAI | 1400 | 5,212 | ±9.3 | 57.3 | $0.50 | |
| 94 | GPT-4.1OpenAI | 1399 | 3,234 | ±10.6 | Not benchmarked | $8.00 | |
| 95 | Qwen3.5-35B-A3BQwen | 1399 | 2,384 | ±12.3 | Not benchmarked | $1.30 | |
| 96 | Qwen3.5-FlashQwen | 1398 | 4,805 | ±9.1 | 41.1 | $0.26 | |
| 97 | GPT-6 SolOpenAI | 1396 | 1,198 | ±17.6 | 76.0 | $10.00 | |
| 98 | GLM 4.6VZ.ai | 1393 | 211 | ±40.1 | Not benchmarked | $0.90 | |
| 99 | Nemotron 3 SuperNVIDIA | 1392 | 552 | ±24.3 | Not benchmarked | $0.45 | |
| 100 | Qwen3 30B A3B Instruct 2507Qwen | 1387 | 1,463 | ±15.4 | Not benchmarked | $0.30 | |
| 101 | Grok 4.3xAI | 1386 | 5,799 | ±8.7 | 37.4 | $2.50 | |
| 102 | Mistral Medium 3Mistral AI | 1384 | 2,119 | ±12.8 | Not benchmarked | $2.00 | |
| 103 | Solar Pro 4Upstage | 1383 | 1,946 | ±14.2 | Not benchmarked | $0.60 | |
| 104 | GLM 4.5 AirZ.ai | 1382 | 1,908 | ±13.4 | Not benchmarked | $1.10 | |
| 105 | Gemma 3 12BGoogle | 1381 | 217 | ±38.2 | Not benchmarked | $0.1625 | |
| 106 | Gemma 3 27BGoogle | 1375 | 2,661 | ±11.6 | Not benchmarked | $0.30 | |
| 107 | GPT-5 MiniOpenAI | 1375 | 1,692 | ±14.3 | 34.2 | $2.00 | |
| 108 | Claude Opus 4RetiredAnthropic | 1373 | 2,850 | ±11.3 | Not benchmarked | $75.00 | |
| 109 | Gemini 2.0 FlashRetiredGoogle | 1371 | 2,559 | ±11.8 | Not benchmarked | $0.40 | |
| 110 | R1RetiredDeepSeek | 1371 | 1,097 | ±17.6 | Not benchmarked | $2.50 | |
| 111 | o1OpenAI | 1368 | 1,454 | ±15.7 | Not benchmarked | $60.00 | |
| 112 | GPT-5.4 NanoOpenAI | 1368 | 4,972 | ±9.3 | 44.8 | $1.25 | |
| 113 | MiniMax M2.5MiniMax | 1366 | 3,299 | ±11.0 | 37.2 | $1.20 | |
| 114 | Qwen3 Next 80B A3B ThinkingQwen | 1364 | 841 | ±20.2 | Not benchmarked | $1.20 | |
| 115 | Nova 2 LiteAmazon | 1359 | 900 | ±19.9 | Not benchmarked | $2.50 | |
| 116 | Qwen3 Coder 480B A35BQwen | 1358 | 1,559 | ±15.4 | Not benchmarked | $1.50 | |
| 117 | GLM 4.7 FlashZ.ai | 1357 | 869 | ±19.3 | Not benchmarked | $0.40 | |
| 118 | gpt-oss-120bOpenAI | 1355 | 1,875 | ±13.7 | Not benchmarked | $0.50 | |
| 119 | o4 MiniOpenAI | 1350 | 2,748 | ±11.5 | Not benchmarked | $4.40 | |
| 120 | Qwen-PlusQwen | 1349 | 392 | ±28.5 | Not benchmarked | $0.78 | |
| 121 | MiniMax M2MiniMax | 1349 | 433 | ±27.4 | Not benchmarked | $1.02 | |
| 122 | Command ACohere | 1348 | 3,432 | ±10.3 | Not benchmarked | $10.00 | |
| 123 | Trinity Large ThinkingArcee AI | 1345 | 2,386 | ±12.8 | Not benchmarked | $0.80 | |
| 124 | GPT-4.1 MiniOpenAI | 1343 | 2,435 | ±12.1 | Not benchmarked | $1.60 | |
| 125 | Nemotron 3 Nano 30B A3BNVIDIA | 1341 | 1,222 | ±16.6 | Not benchmarked | $0.22 | |
| 126 | GLM 4.5VZ.ai | 1338 | 271 | ±33.8 | Not benchmarked | $1.80 | |
| 127 | MiniMax M1MiniMax | 1337 | 2,337 | ±12.3 | Not benchmarked | $2.20 | |
| 128 | o3 MiniOpenAI | 1328 | 1,023 | ±18.0 | Not benchmarked | $4.40 | |
| 129 | Gemini 1.5 ProRetiredGoogle | 1328 | 3,342 | ±11.2 | Not benchmarked | $5.00 | |
| 130 | Gemma 3 4BGoogle | 1327 | 245 | ±35.7 | Not benchmarked | $0.075 | |
| 131 | Granite 4.2 8BIBM Granite | 1327 | 273 | ±36.9 | Not benchmarked | $0.20 | |
| 132 | Qwen3 32BQwen | 1324 | 285 | ±32.0 | Not benchmarked | $0.40 | |
| 133 | o1-miniRetiredOpenAI | 1315 | 3,138 | ±11.7 | Not benchmarked | $4.40 | |
| 134 | Qwen3 30B A3BQwen | 1314 | 1,753 | ±13.9 | Not benchmarked | $0.50 | |
| 135 | GPT-4o (2024-05-13)OpenAI | 1314 | 6,743 | ±10.1 | Not benchmarked | $15.00 | |
| 136 | GPT-4o-mini (2024-07-18)OpenAI | 1305 | 3,866 | ±10.8 | Not benchmarked | $0.60 | |
| 137 | Gemini 1.5 FlashRetiredGoogle | 1302 | 2,160 | ±13.4 | Not benchmarked | $0.30 | |
| 138 | Claude 3.7 SonnetRetiredAnthropic | 1300 | 2,476 | ±12.5 | Not benchmarked | $15.00 | |
| 139 | GPT-4.1 NanoOpenAI | 1300 | 382 | ±28.1 | Not benchmarked | $0.40 | |
| 140 | GPT-5 NanoOpenAI | 1298 | 469 | ±27.0 | Not benchmarked | $0.40 | |
| 141 | Qwen2.5 72B InstructQwen | 1288 | 2,386 | ±12.8 | Not benchmarked | $0.40 | |
| 142 | Llama 4 MaverickMeta Llama | 1286 | 2,582 | ±12.1 | Not benchmarked | $1.00 | |
| 143 | Llama 4 ScoutMeta Llama | 1285 | 1,976 | ±13.8 | Not benchmarked | $0.50 | |
| 144 | Llama 3.3 70B InstructMeta Llama | 1284 | 3,153 | ±10.8 | Not benchmarked | $0.72 | |
| 145 | Llama 3.1 70B InstructMeta Llama | 1283 | 3,177 | ±11.8 | Not benchmarked | $0.56 | |
| 146 | Nova Pro 1.0Amazon | 1281 | 1,296 | ±16.2 | Not benchmarked | $3.20 | |
| 147 | gpt-oss-20bOpenAI | 1280 | 678 | ±23.4 | Not benchmarked | $0.15 | |
| 148 | Claude 3.5 SonnetRetiredAnthropic | 1277 | 5,493 | ±9.0 | Not benchmarked | $15.00 | |
| 149 | GPT-4o (2024-08-06)OpenAI | 1275 | 2,674 | ±12.5 | Not benchmarked | $10.00 | |
| 150 | Mistral LargeMistral AI | 1275 | 1,593 | ±15.0 | Not benchmarked | $6.00 | |
| 151 | Mistral Small 3.1 24BMistral AI | 1271 | 2,115 | ±13.1 | Not benchmarked | $0.555 | |
| 152 | Command R (08-2024)Cohere | 1268 | 592 | ±23.5 | Not benchmarked | $0.60 | |
| 153 | GPT-4 TurboOpenAI | 1265 | 5,381 | ±11.1 | Not benchmarked | $30.00 | |
| 154 | Claude 3 OpusRetiredAnthropic | 1264 | 11,095 | ±9.4 | Not benchmarked | $75.00 | |
| 155 | Claude 3.5 HaikuRetiredAnthropic | 1259 | 4,230 | ±9.8 | Not benchmarked | $4.00 | |
| 156 | Qwen2.5 Coder 32B InstructQwen | 1257 | 273 | ±34.1 | Not benchmarked | $1.00 | |
| 157 | Gemma 2 27BGoogle | 1241 | 4,458 | ±10.4 | Not benchmarked | $0.65 | |
| 158 | Phi 4Microsoft | 1241 | 1,341 | ±16.6 | Not benchmarked | $0.14 | |
| 159 | Nova Lite 1.0Amazon | 1238 | 1,079 | ±18.3 | Not benchmarked | $0.24 | |
| 160 | Nova Micro 1.0Amazon | 1221 | 1,082 | ±18.7 | Not benchmarked | $0.14 | |
| 161 | Mistral Small 3Mistral AI | 1219 | 869 | ±20.7 | Not benchmarked | $0.08 | |
| 162 | Claude 3 SonnetRetiredAnthropic | 1214 | 6,062 | ±11.8 | Not benchmarked | $15.00 | |
| 163 | Llama 3.1 8B InstructMeta Llama | 1206 | 2,882 | ±12.3 | Not benchmarked | $0.08 | |
| 164 | GPT-4OpenAI | 1191 | 2,897 | ±14.2 | Not benchmarked | $60.00 | |
| 165 | Claude 3 HaikuRetiredAnthropic | 1191 | 6,799 | ±10.8 | Not benchmarked | $1.25 | |
| 166 | Llama 3.2 3B InstructMeta Llama | 1115 | 506 | ±27.9 | Not benchmarked | $0.3325 | |
| 167 | Llama 3.2 1B InstructMeta Llama | 1071 | 489 | ±29.0 | Not benchmarked | $0.201 |
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.