LMArena Text: medicine and healthcare leaderboard
- Gemini 4 Argon leads
- 1555
- models measured
- 156
- lead over #2
- +23 Elo
- median
- 1424
Leader on LMArena Text: medicine and healthcare · read October 21
On October 21, 2026, 156 of our catalog's models had a LMArena Text: medicine and healthcare result, and Google Gemini 4 Argon led with 1555.
Prompts from medicine and healthcare. 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
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 ArgonGoogle | 1555 | 332 | ±32.9 | 80.5 | $10.00 | |
| 2 | Claude Fable 5.1Anthropic | 1531 | 1,206 | ±17.6 | 89.1 | $50.00 | |
| 3 | Claude Opus 5Anthropic | 1525 | 2,545 | ±12.6 | 81.8 | $25.00 | |
| 4 | Muse WonderMeta | 1512 | 363 | ±30.2 | 64.5 | $20.00 | |
| 5 | Claude Opus 5.5Anthropic | 1509 | 654 | ±24.1 | 92.2 | $20.00 | |
| 6 | Gemini 3.8 FlashGoogle | 1508 | 2,962 | ±11.8 | 70.5 | $4.125 | |
| 7 | Claude Opus 4.6Anthropic | 1504 | 6,225 | ±8.4 | 66.9 | $25.00 | |
| 8 | Qwen3.8 Max (0902)Qwen | 1503 | 1,991 | ±13.9 | 63.5 | $6.00 | |
| 9 | MiMo-V2.6-ProXiaomi | 1497 | 252 | ±36.2 | 72.6 | $0.87 | |
| 10 | Muse Spark 1.2Meta | 1496 | 660 | ±23.4 | 65.1 | $4.25 | |
| 11 | Claude Opus 4.7Anthropic | 1495 | 4,957 | ±9.2 | 51.1 | $25.00 | |
| 12 | Gemini 3.7 FlashGoogle | 1494 | 2,096 | ±13.6 | 69.9 | $3.75 | |
| 13 | Muse Spark 1.3Meta | 1493 | 1,352 | ±16.4 | 79.7 | $4.25 | |
| 14 | Kimi K3Moonshot AI | 1489 | 2,339 | ±12.7 | 61.8 | $11.936 | |
| 15 | Gemini 3.1 Pro PreviewGoogle | 1486 | 9,660 | ±7.1 | 35.3 | $12.00 | |
| 16 | DeepSeek V4.1 FlashDeepSeek | 1484 | 910 | ±20.1 | 77.2 | $0.525 | |
| 17 | Gemini 3.5 FlashGoogle | 1483 | 3,555 | ±10.6 | 42.8 | $9.00 | |
| 18 | GLM 5.3Z.ai | 1481 | 1,783 | ±14.5 | 73.1 | $1.40 | |
| 19 | Claude Fable 5Anthropic | 1480 | 3,084 | ±11.4 | 82.1 | $50.00 | |
| 20 | Muse Spark 1.1Meta | 1479 | 3,270 | ±11.0 | 48.9 | $4.25 | |
| 21 | Gemini 2.5 ProGoogle | 1476 | 7,991 | ±7.1 | 37.0 | $10.00 | |
| 22 | GLM 5.2Z.ai | 1475 | 3,768 | ±10.3 | 57.7 | $7.00 | |
| 23 | GLM 5.3 FlashZ.ai | 1475 | 2,450 | ±12.4 | 71.0 | $0.50 | |
| 24 | GLM 5.1Z.ai | 1474 | 4,587 | ±9.3 | 44.0 | $4.40 | |
| 25 | MiMo-V2.5-ProXiaomi | 1473 | 5,449 | ±8.8 | 40.0 | $0.609 | |
| 26 | Gemini 3.6 FlashGoogle | 1469 | 2,553 | ±12.2 | 48.3 | $3.75 | |
| 27 | GLM 4.7Z.ai | 1469 | 826 | ±20.1 | 33.6 | $1.98 | |
| 28 | DeepSeek V3.2 ExpDeepSeek | 1468 | 443 | ±27.4 | 55.0 | $0.41 | |
| 29 | DeepSeek V3.1 TerminusDeepSeek | 1468 | 227 | ±37.9 | Not benchmarked | $1.00 | |
| 30 | DeepSeek V4 Pro 0423DeepSeek | 1466 | 4,058 | ±10.0 | 35.3 | $1.9105 | |
| 31 | GPT-5.5OpenAI | 1465 | 5,318 | ±9.0 | 67.8 | $30.00 | |
| 32 | Qwen3.7 PlusQwen | 1464 | 2,965 | ±11.4 | Not benchmarked | $1.28 | |
| 33 | Claude Sonnet 5.5Anthropic | 1464 | 580 | ±24.5 | 88.9 | $10.00 | |
| 34 | Claude Opus 4.8Anthropic | 1464 | 4,758 | ±9.3 | 66.0 | $25.00 | |
| 35 | GPT-5.4OpenAI | 1458 | 4,832 | ±9.3 | 37.7 | $15.00 | |
| 36 | Kimi K2.6Moonshot AI | 1457 | 2,932 | ±11.4 | 44.2 | $3.41 | |
| 37 | GPT-5.1OpenAI | 1456 | 2,627 | ±11.7 | 53.3 | $10.00 | |
| 38 | Qwen3 235B A22B Thinking 2507Qwen | 1456 | 509 | ±25.9 | Not benchmarked | $3.50 | |
| 39 | Nemotron 3 UltraNVIDIA | 1456 | 860 | ±19.9 | Not benchmarked | $2.40 | |
| 40 | GLM 5Z.ai | 1455 | 2,138 | ±12.9 | 59.7 | $3.20 | |
| 41 | Qwen3.5 397B A17BQwen | 1455 | 7,090 | ±7.7 | 28.6 | $2.34 | |
| 42 | Qwen3 Next 80B A3B InstructQwen | 1455 | 1,181 | ±17.2 | Not benchmarked | $1.10 | |
| 43 | GLM 4.5Z.ai | 1455 | 1,386 | ±16.3 | 64.2 | $2.20 | |
| 44 | GLM 4.6Z.ai | 1455 | 1,932 | ±13.4 | 52.2 | $2.20 | |
| 45 | Hy3Tencent | 1452 | 821 | ±20.7 | 43.9 | $0.528 | |
| 46 | Claude Sonnet 4.6Anthropic | 1451 | 5,434 | ±8.8 | 57.8 | $15.00 | |
| 47 | Gemma 4 31BGoogle | 1451 | 422 | ±27.9 | 38.3 | $0.40 | |
| 48 | Mistral Medium 3.1Mistral AI | 1449 | 5,997 | ±8.0 | Not benchmarked | $2.00 | |
| 49 | DeepSeek V4 Flash 0423DeepSeek | 1448 | 3,677 | ±10.4 | 54.6 | $0.182 | |
| 50 | Kimi K2.5Moonshot AI | 1448 | 5,167 | ±8.7 | 58.5 | $2.50 | |
| 51 | o3OpenAI | 1448 | 3,325 | ±11.0 | 70.7 | $8.00 | |
| 52 | T InklingThinking Machines | 1447 | 2,876 | ±11.6 | 49.7 | $4.05 | |
| 53 | DeepSeek V4 Pro 0813DeepSeek | 1446 | 623 | ±23.1 | 54.6 | $1.98 | |
| 54 | R1 0528DeepSeek | 1445 | 1,228 | ±17.5 | 71.4 | $2.15 | |
| 55 | Claude Sonnet 4.5Anthropic | 1444 | 5,397 | ±8.6 | 59.1 | $15.00 | |
| 56 | Claude Opus 4.5Anthropic | 1444 | 4,918 | ±8.9 | 66.8 | $25.00 | |
| 57 | Claude Sonnet 5Anthropic | 1442 | 3,677 | ±10.4 | 63.5 | $10.00 | |
| 58 | Qwen3.6 PlusQwen | 1442 | 3,473 | ±10.6 | 37.4 | $1.95 | |
| 59 | MiniMax M3MiniMax | 1441 | 4,432 | ±9.6 | 57.8 | $1.20 | |
| 60 | DeepSeek V3.2DeepSeek | 1440 | 3,116 | ±10.9 | 54.4 | $0.42 | |
| 61 | Grok 4.5xAI | 1440 | 2,932 | ±11.6 | 75.5 | $6.00 | |
| 62 | MiMo-V2.6-FlashXiaomi | 1439 | 822 | ±20.6 | 61.9 | $0.28 | |
| 63 | Step 5 PreviewStepFun | 1439 | 614 | ±23.9 | 71.4 | $2.70 | |
| 64 | Gemini 3.5 Flash LiteGoogle | 1438 | 2,599 | ±12.2 | 33.6 | $2.50 | |
| 65 | Qwen3.8 27BQwen | 1437 | 1,622 | ±15.2 | 73.6 | $1.35 | |
| 66 | GPT-6.1 SolOpenAI | 1433 | 541 | ±25.2 | 88.7 | $10.00 | |
| 67 | GLM 5V TurboZ.ai | 1433 | 743 | ±21.7 | 28.8 | $4.00 | |
| 68 | Mistral Large 4Mistral AI | 1433 | 678 | ±22.5 | 55.3 | $4.18 | |
| 69 | MiMo-V2.5Xiaomi | 1432 | 3,419 | ±10.8 | 34.0 | $0.238 | |
| 70 | GPT-5.6 SolOpenAI | 1431 | 3,020 | ±11.5 | 76.7 | $20.00 | |
| 71 | Qwen3 VL 235B A22B InstructQwen | 1431 | 627 | ±24.5 | Not benchmarked | $1.90 | |
| 72 | GPT-5.6 TerraOpenAI | 1430 | 3,231 | ±11.1 | 69.0 | $12.00 | |
| 73 | Gemini 2.5 FlashGoogle | 1428 | 7,894 | ±7.0 | 28.7 | $2.50 | |
| 74 | Gemma 4 26B A4BGoogle | 1428 | 391 | ±28.6 | 22.8 | $0.255 | |
| 75 | Gemini 3.1 Flash Lite PreviewGoogle | 1428 | 4,768 | ±9.1 | 8.1 | $1.50 | |
| 76 | Qwen3.5-27BQwen | 1425 | 2,038 | ±13.2 | 22.6 | $2.16 | |
| 77 | GPT-6 AstraOpenAI | 1425 | 958 | ±19.4 | 88.7 | $50.00 | |
| 78 | GPT-5.2OpenAI | 1425 | 6,175 | ±8.2 | 58.9 | $14.00 | |
| 79 | Qwen3.5-122B-A10BQwen | 1423 | 2,185 | ±12.8 | 22.9 | $2.08 | |
| 80 | Qwen3.5-35B-A3BQwen | 1421 | 2,284 | ±12.6 | 7.7 | $1.00 | |
| 81 | Grok 4.6xAI | 1420 | 2,180 | ±13.2 | 76.4 | $6.00 | |
| 82 | Claude Opus 4.1Anthropic | 1416 | 2,704 | ±11.5 | 27.2 | $75.00 | |
| 83 | T Inkling SmallThinking Machines | 1413 | 2,479 | ±12.6 | 29.9 | $1.20 | |
| 84 | GPT-5OpenAI | 1413 | 1,783 | ±14.6 | 72.2 | $10.00 | |
| 85 | Mistral Medium 3.5Mistral AI | 1409 | 844 | ±20.8 | 9.2 | $7.50 | |
| 86 | MiniMax M2.7MiniMax | 1409 | 5,935 | ±8.5 | 28.4 | $1.20 | |
| 87 | Step 3.5 FlashStepFun | 1408 | 4,242 | ±9.5 | Not benchmarked | $0.30 | |
| 88 | Qwen3.5-FlashQwen | 1407 | 4,380 | ±9.5 | 6.2 | $0.26 | |
| 89 | Hy3 previewTencent | 1405 | 481 | ±27.6 | 22.4 | $0.60 | |
| 90 | GLM 4.5 AirZ.ai | 1402 | 1,645 | ±14.4 | Not benchmarked | $0.85 | |
| 91 | GPT-5.6 LunaOpenAI | 1399 | 3,036 | ±11.5 | 61.5 | $1.20 | |
| 92 | Grok 4.7xAI | 1395 | 972 | ±19.3 | 75.0 | $6.00 | |
| 93 | GPT-5.4 MiniOpenAI | 1395 | 4,757 | ±9.4 | 28.2 | $4.50 | |
| 94 | Claude Haiku 4.5Anthropic | 1394 | 10,199 | ±6.7 | 49.8 | $5.00 | |
| 95 | Qwen3 30B A3B Instruct 2507Qwen | 1392 | 1,327 | ±16.4 | Not benchmarked | $0.30 | |
| 96 | Qwen3 Next 80B A3B ThinkingQwen | 1392 | 708 | ±22.3 | Not benchmarked | $1.20 | |
| 97 | Gemma 3 27BGoogle | 1391 | 2,342 | ±12.5 | 4.9 | $0.35 | |
| 98 | GLM 4.6VZ.ai | 1389 | 188 | ±41.4 | Not benchmarked | $0.90 | |
| 99 | gpt-oss-120bOpenAI | 1388 | 1,626 | ±15.0 | 34.5 | $0.60 | |
| 100 | Nova 2 LiteAmazon | 1386 | 746 | ±21.6 | Not benchmarked | $2.50 | |
| 101 | Grok 4.3xAI | 1384 | 5,319 | ±9.0 | 22.5 | $2.50 | |
| 102 | Nemotron 3 SuperNVIDIA | 1383 | 512 | ±25.0 | Not benchmarked | $2.40 | |
| 103 | Mistral Medium 3Mistral AI | 1381 | 1,972 | ±13.6 | Not benchmarked | $2.00 | |
| 104 | GPT-5 MiniOpenAI | 1380 | 1,436 | ±15.8 | 59.8 | $2.00 | |
| 105 | GPT-4.1OpenAI | 1379 | 2,771 | ±11.6 | 46.5 | $8.00 | |
| 106 | R Beam 1.0 501B A23BReflection | 1376 | 311 | ±34.5 | 33.8 | $1.00 | |
| 107 | Solar Pro 4Upstage | 1374 | 1,558 | ±15.7 | 24.4 | $0.60 | |
| 108 | GPT-5.4 NanoOpenAI | 1369 | 4,783 | ±9.5 | Not benchmarked | $1.25 | |
| 109 | Qwen3 32BQwen | 1368 | 212 | ±38.4 | 40.0 | $0.28 | |
| 110 | MiniMax M2MiniMax | 1367 | 315 | ±33.1 | 45.3 | $1.02 | |
| 111 | GPT-6 SolOpenAI | 1366 | 1,051 | ±19.2 | 82.8 | $10.00 | |
| 112 | Mercury 2Inception | 1363 | 231 | ±37.2 | 0.0 | $0.75 | |
| 113 | o4 MiniOpenAI | 1361 | 2,490 | ±12.3 | 68.6 | $4.40 | |
| 114 | MiniMax M2.5MiniMax | 1357 | 3,226 | ±11.1 | 59.4 | $0.95 | |
| 115 | MiniMax M1MiniMax | 1355 | 2,086 | ±13.1 | Not benchmarked | $2.20 | |
| 116 | o1OpenAI | 1355 | 1,273 | ±17.0 | 61.7 | $60.00 | |
| 117 | GPT-6 LunaOpenAI | 1352 | 4,547 | ±10.1 | 71.4 | $0.50 | |
| 118 | Qwen-PlusQwen | 1350 | 315 | ±31.9 | Not benchmarked | $0.78 | |
| 119 | Trinity Large ThinkingArcee AI | 1349 | 2,287 | ±13.2 | 5.4 | $0.80 | |
| 120 | GLM 4.5VZ.ai | 1349 | 233 | ±36.7 | Not benchmarked | $1.80 | |
| 121 | Nemotron 3 Nano 30B A3BNVIDIA | 1349 | 1,015 | ±19.1 | Not benchmarked | $0.24 | |
| 122 | Qwen3 Coder 480B A35BQwen | 1347 | 1,487 | ±16.1 | 50.0 | $1.00 | |
| 123 | GLM 4.7 FlashZ.ai | 1343 | 837 | ±20.4 | Not benchmarked | $0.40 | |
| 124 | Gemma 3 4BGoogle | 1342 | 197 | ±38.2 | Not benchmarked | $0.05 | |
| 125 | Granite 4.2 8BIBM Granite | 1336 | 235 | ±40.2 | Not benchmarked | $0.25 | |
| 126 | GPT-5 NanoOpenAI | 1335 | 469 | ±27.0 | 34.8 | $0.40 | |
| 127 | Command ACohere | 1334 | 3,057 | ±11.0 | Not benchmarked | $10.00 | |
| 128 | Gemma 3 12BGoogle | 1334 | 155 | ±46.7 | Not benchmarked | $0.15 | |
| 129 | o3 MiniOpenAI | 1330 | 824 | ±21.0 | 52.1 | $4.40 | |
| 130 | GPT-4.1 MiniOpenAI | 1324 | 2,308 | ±12.7 | 28.5 | $1.60 | |
| 131 | Qwen3 30B A3BQwen | 1318 | 1,613 | ±14.9 | Not benchmarked | $0.50 | |
| 132 | gpt-oss-20bOpenAI | 1310 | 564 | ±25.3 | Not benchmarked | $0.09 | |
| 133 | GPT-4o-mini (2024-07-18)OpenAI | 1282 | 3,439 | ±11.3 | 3.6 | $0.60 | |
| 134 | GPT-4o (2024-05-13)OpenAI | 1278 | 5,984 | ±10.7 | Not benchmarked | $15.00 | |
| 135 | GPT-4.1 NanoOpenAI | 1276 | 302 | ±32.2 | 8.9 | $0.40 | |
| 136 | Llama 4 MaverickMeta Llama | 1275 | 2,292 | ±13.2 | 18.1 | $0.6525 | |
| 137 | Llama 3.3 70B InstructMeta Llama | 1275 | 2,730 | ±11.9 | Not benchmarked | $0.32 | |
| 138 | Llama 4 ScoutMeta Llama | 1275 | 1,814 | ±14.7 | 9.1 | $0.30 | |
| 139 | Mistral Small 3.1 24BMistral AI | 1267 | 1,946 | ±14.3 | Not benchmarked | $0.555 | |
| 140 | Nova Pro 1.0Amazon | 1260 | 1,096 | ±17.7 | Not benchmarked | $3.20 | |
| 141 | Qwen2.5 72B InstructQwen | 1258 | 1,956 | ±14.0 | Not benchmarked | $0.40 | |
| 142 | Llama 3.1 70B InstructMeta Llama | 1257 | 2,821 | ±12.6 | Not benchmarked | $0.40 | |
| 143 | Mistral LargeMistral AI | 1255 | 1,230 | ±16.7 | Not benchmarked | $6.00 | |
| 144 | GPT-4o (2024-08-06)OpenAI | 1252 | 2,174 | ±13.8 | 23.1 | $10.00 | |
| 145 | GPT-4 TurboOpenAI | 1239 | 5,046 | ±11.6 | Not benchmarked | $30.00 | |
| 146 | Command R (08-2024)Cohere | 1238 | 547 | ±24.5 | Not benchmarked | $0.60 | |
| 147 | Mistral Small 3Mistral AI | 1215 | 654 | ±23.7 | Not benchmarked | $0.08 | |
| 148 | Nova Lite 1.0Amazon | 1213 | 872 | ±19.9 | Not benchmarked | $0.24 | |
| 149 | Nova Micro 1.0Amazon | 1207 | 952 | ±19.1 | Not benchmarked | $0.14 | |
| 150 | Phi 4Microsoft | 1202 | 1,037 | ±19.4 | Not benchmarked | $0.14 | |
| 151 | Gemma 2 27BGoogle | 1198 | 3,862 | ±11.3 | Not benchmarked | $0.65 | |
| 152 | Qwen2.5 Coder 32B InstructQwen | 1187 | 221 | ±36.4 | 30.5 | $1.00 | |
| 153 | Llama 3.1 8B InstructMeta Llama | 1182 | 2,564 | ±13.1 | Not benchmarked | $0.08 | |
| 154 | GPT-4OpenAI | 1145 | 2,913 | ±14.7 | 22.4 | $60.00 | |
| 155 | Llama 3.2 3B InstructMeta Llama | 1104 | 441 | ±30.0 | Not benchmarked | $0.335 | |
| 156 | Llama 3.2 1B InstructMeta Llama | 1009 | 445 | ±31.1 | Not benchmarked | $0.201 |
Only models in our catalog are listed, and each appears once with its best published run. 172 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.