Instruction Following
LMArena Text Leaderboard
Arena (arena.ai, formerly LMArena / Chatbot Arena) ranks models by millions of blind, head-to-head human votes on real prompts. The Arena Score shown here is a Bradley-Terry rating computed with Arena's default style control, so higher means people preferred that model's answers more often — it's a relative rating, not a percentage, so compare it only to other models on this board.
Source: lmarena182 open models ranked+209 proprietaryData through Sep 2026
All models ranked on LMArena Text
Proprietary / closed models are shown dimmed — you can't run them locally, but they show where the open field stands.
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 · proprietary | 1505.7 |
| 2 | Claude Opus 4.6 High · proprietary | 1504.6 |
| 3 | Claude Opus 4.7 High · proprietary | 1501.7 |
| 4 | Muse Spark 1.2 (xHigh) · proprietary | 1499.6 |
| 5 | Claude Fable 5.1 Max · proprietary | 1498.5 |
| 6 | Claude Opus 4.6 · proprietary | 1497.3 |
| 7 | Claude Opus 4.7 · proprietary | 1494.4 |
| 8 | Muse Spark 1.3 Max · proprietary | 1493.1 |
| 9 | Gemini 3.8 Flash High · proprietary | 1493.0 |
| 10 | Claude Opus 5 High · proprietary | 1492.9 |
| 11 | Muse Spark 1.1 · proprietary | 1492.6 |
| 12 | Gemini 3.7 Flash High · proprietary | 1489.8 |
| 13 | Muse Spark · proprietary | 1488.1 |
| 14 | Claude Opus 5 Max · proprietary | 1487.4 |
| 15 | Gemini 3.1 Pro Preview · proprietary | 1486.8 |
| 16 | Gemini 3 Pro · proprietary | 1485.5 |
| 17 | Kimi K3 · 2779.9B | 1484.8 |
| 18 | GPT 5.6 Sol Xhigh · proprietary | 1483.5 |
| 19 | GLM 5.3 · 753.3B | 1483.0 |
| 20 | GPT 5.5 High · proprietary | 1481.8 |
| 21 | Claude Opus 4.8 High · proprietary | 1481.4 |
| 22 | Qwen3.8 Max · proprietary | 1480.6 |
| 23 | Gemini 3.6 Flash High · proprietary | 1479.9 |
| 24 | GPT 6 Astra Max · proprietary | 1479.8 |
| 25 | Gemini 3.5 Flash High · proprietary | 1477.7 |
| 26 | GPT 5.4 High · proprietary | 1476.4 |
| 27 | GPT 5.2 Chat Latest (Feb 10, 2026) · proprietary | 1476.3 |
| 28 | GPT 5.5 · proprietary | 1476.1 |
| 29 | GLM 5.3 Flash · 321.3B | 1475.4 |
| 30 | Grok 4.20 Beta1 · proprietary | 1474.5 |
| 31 | Gemini 3.5 Flash Medium · proprietary | 1473.9 |
| 32 | GPT 5.5 Instant · proprietary | 1473.8 |
| 33 | Gemini 3 Flash · proprietary | 1473.7 |
| 34 | Qwen3.7 Max Preview · proprietary | 1473.2 |
| 35 | Claude Opus 4.8 · proprietary | 1473.2 |
| 36 | Claude Opus 4.5 High 32K (Nov 01, 2025) · proprietary | 1472.8 |
| 37 | Claude Sonnet 4.6 · proprietary | 1472.6 |
| 38 | GLM 5.2 · 753.3B | 1472.1 |
| 39 | Grok 4.20 Beta 0309 Reasoning · proprietary | 1471.5 |
| 40 | Grok 4.20 Multi Agent Beta (Mar 09) · proprietary | 1470.2 |
| 41 | Claude Opus 4.5 (Nov 01, 2025) · proprietary | 1469.5 |
| 42 | Grok 4.5 · proprietary | 1468.1 |
| 43 | Ernie 5.1 · proprietary | 1467.9 |
| 44 | MiMo V2.5 Pro · 1023.2B | 1467.4 |
| 45 | GPT 5.6 Terra Xhigh · proprietary | 1466.0 |
| 46 | GPT 5.4 · proprietary | 1465.7 |
| 47 | GLM 5.1 · 753.9B | 1465.5 |
| 48 | Grok 4.1 Thinking · proprietary | 1465.4 |
| 49 | Qwen3.5 Max Preview · proprietary | 1465.2 |
| 50 | DeepSeek V4 Pro 0813 · 1650.5B | 1463.4 |
| 51 | Claude Sonnet 5 High · proprietary | 1461.2 |
| 52 | Kimi K2.6 · 1026.9B | 1460.4 |
| 53 | Qwen3.6 Max Preview · proprietary | 1460.0 |
| 54 | Grok 4.1 · proprietary | 1459.2 |
| 55 | Gemini 3 Flash (thinking Minimal) · proprietary | 1458.3 |
| 56 | GLM 5 · 753.9B | 1457.6 |
| 57 | DeepSeek V4 Pro · 1598.8B | 1457.3 |
| 58 | Claude Sonnet 4.5 High 32K (Sep 29, 2025) · proprietary | 1456.2 |
| 59 | Gemini 3.5 Flash Lite · proprietary | 1455.9 |
| 60 | Hy3 · 298.8B | 1455.8 |
| 61 | Dola Seed 2.0 Pro · proprietary | 1455.6 |
| 62 | Qwen3.7 Plus · proprietary | 1455.6 |
| 63 | Grok 4.6 High · proprietary | 1455.6 |
| 64 | GPT 5.1 High · proprietary | 1455.1 |
| 65 | Claude Sonnet 4.5 (Sep 29, 2025) · proprietary | 1455.1 |
| 66 | GPT 5.6 Luna Xhigh · proprietary | 1452.5 |
| 67 | Gemma 4 31B IT · 31.3B | 1451.1 |
| 68 | Kimi K2.5 · 1026.9B | 1450.4 |
| 69 | Claude Opus 4.1 Thinking 16K (Aug 05, 2025) · proprietary | 1449.9 |
| 70 | GPT 5.3 Chat Latest · proprietary | 1449.2 |
| 71 | Ernie 5.0 Preview (Dec 03) · proprietary | 1448.9 |
| 72 | Mimo v2 Pro · proprietary | 1448.4 |
| 73 | GPT 5.4 Mini High · proprietary | 1448.1 |
| 74 | Claude Opus 4.1 (Aug 05, 2025) · proprietary | 1447.7 |
| 75 | Ernie 5.0 (Jan 10) · proprietary | 1446.1 |
| 76 | Gemini 2.5 Pro · proprietary | 1445.5 |
| 77 | GPT 4.5 Preview (Feb 27, 2025) · proprietary | 1444.6 |
| 78 | Qwen3.6 Plus · proprietary | 1443.3 |
| 79 | ChatGPT 4o Latest (Mar 26, 2025) · proprietary | 1443.1 |
| 80 | Grok 4.3 · proprietary | 1442.7 |
| 81 | Qwen3.5 397B A17B · 403.4B | 1441.8 |
| 82 | GLM 4.7 · 358.3B | 1441.7 |
| 83 | MiniMax M3 · 427.0B | 1441.3 |
| 84 | Inkling · 952.4B | 1440.1 |
| 85 | GPT 5.1 · proprietary | 1439.0 |
| 86 | DeepSeek V4 Flash · 290.9B | 1437.9 |
| 87 | Gemma 4 26B A4B IT · 25.8B | 1437.9 |
| 88 | GPT 5.2 High · proprietary | 1437.6 |
| 89 | Qwen3.8 27B · 27.8B | 1437.1 |
| 90 | GPT 5.2 · proprietary | 1435.8 |
| 91 | Longcat Flash Chat 2602 Exp · proprietary | 1435.5 |
| 92 | Qwen3 Max Preview · proprietary | 1434.7 |
| 93 | GPT 5 High · proprietary | 1434.7 |
| 94 | MiMo V2.5 · 310.8B | 1433.6 |
| 95 | GLM 5v Turbo · proprietary | 1432.7 |
| 96 | Gemini 3.1 Flash Lite Preview · proprietary | 1432.4 |
| 97 | O3 (Apr 16, 2025) · proprietary | 1431.6 |
| 98 | Grok 4.1 Fast Reasoning · proprietary | 1430.3 |
| 99 | Kimi K2 Thinking · 1026.4B | 1430.1 |
| 100 | Mimo v2 Omni · proprietary | 1429.8 |
| 101 | Muse Glimmer 30B · 29.8B | 1427.2 |
| 102 | GPT 5 Chat · proprietary | 1426.8 |
| 103 | Mistral Medium 3.5 128B · 127.7B | 1426.3 |
| 104 | Claude Opus 4 Thinking 16K (May 14, 2025) · proprietary | 1426.0 |
| 105 | Amazon Nova Experimental Chat 10 (26 02) · proprietary | 1425.9 |
| 106 | NVIDIA Nemotron 3 Ultra 550B A55B BF16 · 560.5B | 1425.6 |
| 107 | DeepSeek V3.2 Exp · 685.4B | 1425.3 |
| 108 | DeepSeek V3.2 · 685.4B | 1425.2 |
| 109 | GLM 4.6 · 356.8B | 1424.7 |
| 110 | Qwen3 Max (Sep 23, 2025) · proprietary | 1423.7 |
| 111 | Qwen3 235B A22B Instruct 2507 · 235.1B | 1423.0 |
| 112 | DeepSeek R1 0528 · 684.5B | 1421.4 |
| 113 | Ernie 5.0 Preview (Oct 22) · proprietary | 1418.8 |
| 114 | Grok 4 Fast Chat · proprietary | 1418.4 |
| 115 | Kimi K2 Instruct 0905 · 1026.5B | 1418.1 |
| 116 | Kimi K2 Instruct · 1026.4B | 1417.8 |
| 117 | DeepSeek V3.1 Terminus · 684.5B | 1417.4 |
| 118 | DeepSeek V3.1 · 684.5B | 1417.3 |
| 119 | Qwen3.5 122B A10B · 125.1B | 1416.8 |
| 120 | MiniMax M2.7 · 228.7B | 1415.2 |
| 121 | Amazon Nova Experimental Chat 10 (26 01) · proprietary | 1414.6 |
| 122 | Claude Haiku 4.5 (Oct 01, 2025) · proprietary | 1414.6 |
| 123 | GPT 4.1 (Apr 14, 2025) · proprietary | 1414.5 |
| 124 | Qwen3 VL 235B A22B Instruct · 235.7B | 1414.1 |
| 125 | Claude Opus 4 (May 14, 2025) · proprietary | 1413.9 |
| 126 | Mistral Large 3 675B Instruct 2512 · 675B | 1413.0 |
| 127 | Hy3 Preview · 298.8B | 1412.7 |
| 128 | Grok 3 Preview (Feb 24) · proprietary | 1411.5 |
| 129 | GLM 4.5 · 358.3B | 1411.3 |
| 130 | Grok 4 (Jul 09) · proprietary | 1411.0 |
| 131 | Gemini 2.5 Flash · proprietary | 1409.7 |
| 132 | Mistral Medium 2508 · proprietary | 1408.6 |
| 133 | Qwen3.5 27B · 27.8B | 1407.9 |
| 134 | Inkling Small · 266.0B | 1404.6 |
| 135 | Grok 4 Fast Reasoning · proprietary | 1404.6 |
| 136 | Gemini 2.5 Flash Preview (Sep 2025) · proprietary | 1403.6 |
| 137 | Qwen3 235B A22B · 235.1B | 1402.5 |
| 138 | O1 (Dec 17, 2024) · proprietary | 1402.1 |
| 139 | GPT 5.4 Nano High · proprietary | 1402.1 |
| 140 | LongCat Flash Chat · 561.9B | 1401.3 |
| 141 | Claude Sonnet 4 Thinking 32K (May 14, 2025) · proprietary | 1400.9 |
| 142 | Qwen3 235B A22B Thinking 2507 · 235.1B | 1400.1 |
| 143 | Qwen3 Next 80B A3B Instruct · 81.3B | 1399.3 |
| 144 | DeepSeek R1 · 684.5B | 1398.0 |
| 145 | Qwen3.5 Flash · proprietary | 1396.7 |
| 146 | DeepSeek v3 0324 · 684.5B | 1395.7 |
| 147 | Qwen3 VL 235B A22B Thinking · 235.7B | 1395.2 |
| 148 | Hunyuan Vision 1.5 Thinking · proprietary | 1395.1 |
| 149 | Qwen3.5 35B A3B · 36.0B | 1395.0 |
| 150 | Amazon Nova Experimental Chat (Dec 10) · proprietary | 1394.6 |
| 151 | Step 3.5 Flash · 199.4B | 1394.1 |
| 152 | MiMo v2 Flash · 309.8B | 1391.9 |
| 153 | O4 Mini (Apr 16, 2025) · proprietary | 1390.8 |
| 154 | Claude Sonnet 4 (May 14, 2025) · proprietary | 1390.7 |
| 155 | MiniMax M2.5 · 228.7B | 1390.3 |
| 156 | GPT 5 Mini High · proprietary | 1389.6 |
| 157 | O1 Preview · proprietary | 1388.4 |
| 158 | Claude 3.7 Sonnet Thinking 32K (Feb 19, 2025) · proprietary | 1388.1 |
| 159 | Qwen3 Coder 480B A35B Instruct · 480.2B | 1387.5 |
| 160 | Hunyuan T1 (Jul 11, 2025) · proprietary | 1387.5 |
| 161 | Mistral Medium 2505 · proprietary | 1387.4 |
| 162 | MiniMax M2.1 · 228.7B | 1383.8 |
| 163 | Hunyuan Turbos (Apr 16, 2025) · proprietary | 1382.6 |
| 164 | GPT 4.1 Mini (Apr 14, 2025) · proprietary | 1382.5 |
| 165 | Qwen3 30B A3B Instruct 2507 · 30.5B | 1382.3 |
| 166 | Gemini 2.5 Flash Lite Preview No Thinking (Sep 2025) · proprietary | 1379.6 |
| 167 | GLM 4.6V · 107.7B | 1378.5 |
| 168 | Trinity Large Preview · 398.6B | 1378.4 |
| 169 | Solar Pro4 · proprietary | 1376.4 |
| 170 | Gemini 2.5 Flash Lite Preview Thinking (Jun 17) · proprietary | 1374.6 |
| 171 | Claude 3.5 Sonnet (Oct 22, 2024) · proprietary | 1374.0 |
| 172 | Qwen2.5 Max · proprietary | 1373.9 |
| 173 | GLM 4.5 Air · 110.5B | 1373.2 |
| 174 | Claude 3.7 Sonnet (Feb 19, 2025) · proprietary | 1372.2 |
| 175 | Qwen3 Next 80B A3B Thinking · 81.3B | 1369.0 |
| 176 | Trinity Large Thinking · 398.6B | 1369.0 |
| 177 | GLM 4.7 Flash · 31.2B | 1366.3 |
| 178 | Gemma 3 27B IT · 27.4B | 1365.2 |
| 179 | Amazon Nova Experimental Chat (Nov 10) · proprietary | 1365.1 |
| 180 | MiniMax M1 · proprietary | 1363.9 |
| 181 | Grok 3 Mini High · proprietary | 1363.5 |
| 182 | O3 Mini High · proprietary | 1363.5 |
| 183 | NVIDIA Nemotron 3 Super 120B A12B BF16 · 123.6B | 1360.5 |
| 184 | Gemini 2.0 Flash 001 · proprietary | 1359.9 |
| 185 | DeepSeek v3 · 684.5B | 1358.4 |
| 186 | Grok 3 Mini Beta · proprietary | 1357.9 |
| 187 | Mistral Small 3.2 24B Instruct 2506 · 24.0B | 1356.7 |
| 188 | INTELLECT 3 · 106.9B | 1356.3 |
| 189 | C4ai Command A 03 2025 · 111.1B | 1353.6 |
| 190 | Gemini 2.0 Flash Lite Preview (Feb 05) · proprietary | 1353.4 |
| 191 | GLM 4.5V · 107.7B | 1352.6 |
| 192 | GPT OSS 120B · 116.8B | 1352.3 |
| 193 | NVIDIA Nemotron 3.5 Lightning 30B A3B BF16 · 31.6B | 1351.5 |
| 194 | Gemini 1.5 Pro 002 · proprietary | 1351.1 |
| 195 | Amazon Nova Experimental Chat (Oct 20) · proprietary | 1349.5 |
| 196 | Step3 · 321.0B | 1349.4 |
| 197 | Hunyuan Turbos (Feb 26, 2025) · proprietary | 1348.8 |
| 198 | O3 Mini · proprietary | 1347.7 |
| 199 | Llama 3 1 Nemotron Ultra 253B V1 · 253.4B | 1347.5 |
| 200 | Qwen3 32B · 32.8B | 1346.9 |
| 201 | Amazon Nova Experimental Chat (Oct 09) · proprietary | 1346.8 |
| 202 | Mercury 2 · proprietary | 1346.3 |
| 203 | GPT 4o (May 13, 2024) · proprietary | 1345.9 |
| 204 | Qwen Plus (Jan 25) · proprietary | 1345.9 |
| 205 | MiniMax M2 · 228.7B | 1345.9 |
| 206 | Ling Flash 2.0 · 102.9B | 1344.0 |
| 207 | Llama 3 3 Nemotron Super 49B V1 5 · 49.9B | 1343.1 |
| 208 | Claude 3.5 Sonnet (Jun 20, 2024) · proprietary | 1343.0 |
| 209 | GLM 4 Plus (Jan 11) · proprietary | 1342.6 |
| 210 | Granite 4.2 30B · 29.3B | 1341.9 |
| 211 | Gemma 3 12B IT · 12.2B | 1341.7 |
| 212 | Hunyuan Turbo (Jan 10) · proprietary | 1340.6 |
| 213 | GPT 5 Nano High · proprietary | 1337.5 |
| 214 | O1 Mini · proprietary | 1336.8 |
| 215 | Nova 2 Lite · proprietary | 1335.9 |
| 216 | QwQ 32B · 32.8B | 1335.8 |
| 217 | Gemini Advanced (May 14) · proprietary | 1335.7 |
| 218 | Grok 2 (Aug 13, 2024) · proprietary | 1335.5 |
| 219 | GPT 4o (Aug 06, 2024) · proprietary | 1335.4 |
| 220 | Llama 3.1 405B Instruct · 405.9B | 1335.2 |
| 221 | Step 2 16K Exp 202412 · proprietary | 1333.5 |
| 222 | Olmo 3.1 32B Instruct · 32.2B | 1330.0 |
| 223 | Yi Lightning · proprietary | 1328.2 |
| 224 | Molmo2 8B · 8.7B | 1328.0 |
| 225 | Llama 3 3 Nemotron Super 49B V1 · 49.9B | 1327.7 |
| 226 | Llama 4 Maverick 17B 128E Instruct · 401.6B | 1326.9 |
| 227 | Qwen3 30B A3B · 30.5B | 1326.9 |
| 228 | Hunyuan Large (Feb 10, 2025) · proprietary | 1326.2 |
| 229 | Claude 3.5 Haiku (Oct 22, 2024) · proprietary | 1324.3 |
| 230 | GPT 4 Turbo (Apr 09, 2024) · proprietary | 1324.3 |
| 231 | Gemini 1.5 Pro 001 · proprietary | 1323.8 |
| 232 | DeepSeek V2.5 1210 · 235.7B | 1323.3 |
| 233 | GPT 4.1 Nano (Apr 14, 2025) · proprietary | 1322.2 |
| 234 | Claude 3 Opus (Feb 29, 2024) · proprietary | 1321.9 |
| 235 | Llama 4 Scout 17B 16E Instruct · 108.6B | 1321.5 |
| 236 | Ring Flash 2.0 · 102.9B | 1320.9 |
| 237 | Step 1o Turbo 202506 · proprietary | 1320.1 |
| 238 | GLM 4 Plus · proprietary | 1319.3 |
| 239 | Qwen Max (Sep 19) · proprietary | 1318.0 |
| 240 | Llama 3.3 70B Instruct · 70.6B | 1317.6 |
| 241 | GPT 4o Mini (Jul 18, 2024) · proprietary | 1317.6 |
| 242 | GPT OSS 20B · 20.9B | 1317.2 |
| 243 | Gemma 3n E4B IT · 7.8B | 1317.2 |
| 244 | Qwen2.5 Plus (Nov 27) · proprietary | 1314.6 |
| 245 | NVIDIA Nemotron 3 Nano 30B A3B BF16 · 31.6B | 1314.2 |
| 246 | Mistral Large Instruct 2407 · 122.6B | 1314.2 |
| 247 | Athene v2 Chat · 72.7B | 1314.1 |
| 248 | GPT 4 0125 Preview · proprietary | 1313.0 |
| 249 | GPT 4 1106 Preview · proprietary | 1312.6 |
| 250 | Hunyuan Standard (Feb 10, 2025) · proprietary | 1311.2 |
| 251 | Gemini 1.5 Flash 002 · proprietary | 1309.1 |
| 252 | Grok 2 Mini (Aug 13, 2024) · proprietary | 1308.3 |
| 253 | Mercury · proprietary | 1308.0 |
| 254 | DeepSeek V2.5 · 235.7B | 1307.1 |
| 255 | Olmo 3 32B Think · 32.2B | 1306.7 |
| 256 | Athene 70B (Jul 25) · proprietary | 1306.4 |
| 257 | Granite 4.1 8B · 8.8B | 1306.0 |
| 258 | Mistral Large Instruct 2411 · 122.6B | 1305.4 |
| 259 | Magistral Medium 2506 · proprietary | 1304.7 |
| 260 | Gemma 3 4B IT · 4.3B | 1303.1 |
| 261 | Mistral Small 3.1 24B Instruct 2503 · 24.0B | 1303.1 |
| 262 | Qwen2.5 72B Instruct · 72.7B | 1302.8 |
| 263 | Llama 3.1 Nemotron 70B Instruct HF · 70.6B | 1298.7 |
| 264 | Hunyuan Large Vision · proprietary | 1294.1 |
| 265 | Granite 4.2 3B · 3.7B | 1293.7 |
| 266 | Llama 3.1 70B Instruct · 70.6B | 1293.1 |
| 267 | Amazon Nova Pro v1.0 · proprietary | 1290.0 |
| 268 | AI21 Jamba Large 1.5 · 398.6B | 1289.3 |
| 269 | Gemma 2 27B IT · 27.2B | 1289.2 |
| 270 | Reka Core (Sep 04, 2024) · proprietary | 1287.8 |
| 271 | GPT 4 (Mar 14) · proprietary | 1287.5 |
| 272 | Granite 4.2 8B · 8.8B | 1286.6 |
| 273 | Llama 3 1 Nemotron 51B Instruct · 51B | 1286.3 |
| 274 | Gemini 1.5 Flash 001 · proprietary | 1286.3 |
| 275 | Llama 3.1 Tulu 3 70B · 70.6B | 1285.9 |
| 276 | Granite 4.0 H Small · 32.2B | 1285.5 |
| 277 | Olmo 3.1 32B Think · 32.2B | 1285.5 |
| 278 | Claude 3 Sonnet (Feb 29, 2024) · proprietary | 1281.1 |
| 279 | Gemma 2 9B IT SimPO · 9.2B | 1280.1 |
| 280 | Nemotron 4 340B Instruct · proprietary | 1276.9 |
| 281 | Meta Llama 3 70B Instruct · 70.6B | 1276.3 |
| 282 | C4ai Command R Plus 08 2024 · 103.8B | 1276.1 |
| 283 | GPT 4 (Jun 13) · proprietary | 1275.8 |
| 284 | Mistral Small 24B Instruct 2501 · 23.6B | 1274.3 |
| 285 | GLM 4 (May 20) · proprietary | 1273.2 |
| 286 | Reka Flash (Sep 04, 2024) · proprietary | 1272.1 |
| 287 | Qwen2.5 Coder 32B Instruct · 32.8B | 1270.5 |
| 288 | Aya Expanse 32B · 32.3B | 1267.0 |
| 289 | Gemma 2 9B IT · 9.2B | 1266.7 |
| 290 | DeepSeek Coder v2 Instruct · 235.7B | 1265.1 |
| 291 | Qwen2 72B Instruct · 72.7B | 1261.5 |
| 292 | C4ai Command R Plus · 103.8B | 1261.5 |
| 293 | Claude 3 Haiku (Mar 07, 2024) · proprietary | 1261.4 |
| 294 | Amazon Nova Lite v1.0 · proprietary | 1260.3 |
| 295 | Gemini 1.5 Flash 8B 001 · proprietary | 1258.6 |
| 296 | Phi 4 · 14.7B | 1256.1 |
| 297 | OLMo 2 0325 32B Instruct · 32.2B | 1251.5 |
| 298 | C4ai Command R 08 2024 · 32.3B | 1250.1 |
| 299 | Mistral Large 2402 · proprietary | 1242.2 |
| 300 | Amazon Nova Micro v1.0 · proprietary | 1240.6 |
| 301 | AI21 Jamba Mini 1.5 · 51.6B | 1239.6 |
| 302 | Ministral 8B Instruct 2410 · 8.0B | 1237.5 |
| 303 | Gemini Pro Dev Api · proprietary | 1236.5 |
| 304 | Qwen1.5 110B Chat · 111.2B | 1233.9 |
| 305 | Reka Flash 21B Online (Feb 26, 2024) · proprietary | 1233.3 |
| 306 | Qwen1.5 72B Chat · 72.3B | 1233.2 |
| 307 | Hunyuan Standard 256K · proprietary | 1233.2 |
| 308 | Mixtral 8x22B Instruct v0.1 · 140.6B | 1229.3 |
| 309 | C4ai Command R V01 · 35.0B | 1226.6 |
| 310 | Reka Flash 21B (Feb 26, 2024) · proprietary | 1226.5 |
| 311 | GPT 3.5 Turbo (Jan 25) · proprietary | 1225.1 |
| 312 | Meta Llama 3 8B Instruct · 8.0B | 1223.4 |
| 313 | Gemini Pro · proprietary | 1223.2 |
| 314 | Aya Expanse 8B · 8.0B | 1222.9 |
| 315 | Mistral Medium · proprietary | 1222.4 |
| 316 | Llama 3.1 Tulu 3 8B · 8.0B | 1220.3 |
| 317 | Zephyr Orpo 141B A35b v0.1 · 140.6B | 1212.7 |
| 318 | Yi 1.5 34B Chat · 34.4B | 1212.6 |
| 319 | Llama 3.1 8B Instruct · 8.0B | 1211.1 |
| 320 | Granite 3.1 8B Instruct · 8.2B | 1208.2 |
| 321 | GPT 3.5 Turbo (Nov 06) · proprietary | 1203.8 |
| 322 | Qwen1.5 32B Chat · 32.5B | 1203.7 |
| 323 | Gemma 2 2B IT · 2.6B | 1199.9 |
| 324 | Phi 3 Medium 4k Instruct · 14.0B | 1197.6 |
| 325 | Mixtral 8x7B Instruct v0.1 · 46.7B | 1196.9 |
| 326 | Dbrx Instruct Preview · proprietary | 1195.2 |
| 327 | Qwen1.5 14B Chat · 14.2B | 1190.8 |
| 328 | Internlm2 5 20B Chat · 19.9B | 1190.6 |
| 329 | Deepseek Llm 67B Chat · 67B | 1184.6 |
| 330 | WizardLM 70B · proprietary | 1184.5 |
| 331 | Yi 34B Chat · 34.4B | 1183.5 |
| 332 | Granite 3.0 8B Instruct · 8.2B | 1182.8 |
| 333 | Openchat 3.5 · proprietary | 1182.7 |
| 334 | Openchat 3.5 0106 · 7.2B | 1182.4 |
| 335 | Gemma 1.1 7B IT · 8.5B | 1182.2 |
| 336 | Snowflake Arctic Instruct · 478.6B | 1179.7 |
| 337 | Granite 3.1 2B Instruct · 2.5B | 1178.6 |
| 338 | Tulu 2 Dpo 70B · proprietary | 1177.4 |
| 339 | OpenHermes 2.5 Mistral 7B · 7B | 1175.6 |
| 340 | Vicuna 33B V1.3 · 33B | 1172.6 |
| 341 | Starling LM 7B Beta · 7.2B | 1170.8 |
| 342 | Phi 3 Small 8k Instruct · 7.4B | 1170.8 |
| 343 | Llama 2 70B Chat HF · 69.0B | 1170.4 |
| 344 | Starling LM 7B Alpha · 7.2B | 1167.1 |
| 345 | Llama 3.2 3B Instruct · 3.2B | 1166.5 |
| 346 | Nous Hermes 2 Mixtral 8x7B DPO · 46.7B | 1164.2 |
| 347 | Granite 3.0 2B Instruct · 2.6B | 1156.3 |
| 348 | Llama2 70B Steerlm Chat · proprietary | 1154.4 |
| 349 | QwQ 32B Preview · 32.8B | 1154.2 |
| 350 | SOLAR 10.7B Instruct v1.0 · 10.7B | 1152.1 |
| 351 | Dolphin 2.2.1 Mistral 7B · proprietary | 1152.0 |
| 352 | Mpt 30B Chat · proprietary | 1150.4 |
| 353 | WizardLM 13B · proprietary | 1149.2 |
| 354 | Mistral 7B Instruct v0.2 · 7.2B | 1149.1 |
| 355 | Falcon 180B Chat · proprietary | 1147.7 |
| 356 | Qwen1.5 7B Chat · 7.7B | 1143.6 |
| 357 | Phi 3 Mini 4k Instruct · 3.8B | 1142.8 |
| 358 | Vicuna 13B · proprietary | 1141.2 |
| 359 | Llama 2 13B Chat HF · 13.0B | 1141.2 |
| 360 | Qwen 14B Chat · 14.2B | 1139.0 |
| 361 | PaLM 2 · proprietary | 1138.7 |
| 362 | Gemma 7B IT · 8.5B | 1137.4 |
| 363 | CodeLlama 34B Instruct HF · 33.7B | 1136.6 |
| 364 | Zephyr 7B Beta · 7.2B | 1130.5 |
| 365 | Phi 3 Mini 128k Instruct · 3.8B | 1129.5 |
| 366 | Guanaco 33B · proprietary | 1127.2 |
| 367 | Zephyr 7B Alpha · proprietary | 1126.6 |
| 368 | StripedHyena Nous 7B · 7.6B | 1121.3 |
| 369 | CodeLlama 70B Instruct HF · 69.0B | 1118.9 |
| 370 | Gemma 1.1 2B IT · 2.5B | 1116.3 |
| 371 | Vicuna 7B · proprietary | 1115.0 |
| 372 | SmolLM2 1.7B Instruct · 1.7B | 1114.4 |
| 373 | Llama 3.2 1B Instruct · 1.2B | 1110.8 |
| 374 | Mistral 7B Instruct v0.1 · 7.2B | 1110.0 |
| 375 | Llama 2 7B Chat HF · 6.7B | 1107.7 |
| 376 | Gemma 2B IT · 2.5B | 1093.3 |
| 377 | Qwen1.5 4B Chat · proprietary | 1090.7 |
| 378 | OLMo 7B Instruct · proprietary | 1073.4 |
| 379 | Koala 13B · proprietary | 1070.6 |
| 380 | Alpaca 13B · proprietary | 1069.7 |
| 381 | Gpt4all 13B Snoozy · proprietary | 1067.3 |
| 382 | Mpt 7B Chat · proprietary | 1063.0 |
| 383 | Chatglm3 6B · 6.2B | 1056.2 |
| 384 | RWKV 4 Raven 14B · proprietary | 1042.0 |
| 385 | Chatglm2 6B · 6B | 1024.4 |
| 386 | Oasst Pythia 12B · proprietary | 1023.1 |
| 387 | Chatglm 6B · proprietary | 995.5 |
| 388 | Fastchat T5 3B · proprietary | 992.2 |
| 389 | Dolly v2 12B · proprietary | 982.0 |
| 390 | Llama 13B · proprietary | 974.6 |
| 391 | Stablelm Tuned Alpha 7B · 7B | 952.9 |
Score vs model size
Which models give the most quality for their size — the ones worth running locally.
- Llama 3.2 1B Instruct, 1B, score 1110.8 — on the efficiency frontier (best score at its size or smaller).
- SmolLM2 1.7B Instruct, 2B, score 1114.4 — on the efficiency frontier (best score at its size or smaller).
- Gemma 1.1 2B IT, 3B, score 1116.3 — on the efficiency frontier (best score at its size or smaller).
- Granite 3.1 2B Instruct, 3B, score 1178.6 — on the efficiency frontier (best score at its size or smaller).
- Gemma 2 2B IT, 3B, score 1199.9 — on the efficiency frontier (best score at its size or smaller).
- Granite 4.2 3B, 4B, score 1293.7 — on the efficiency frontier (best score at its size or smaller).
- Gemma 3 4B IT, 4B, score 1303.1 — on the efficiency frontier (best score at its size or smaller).
- Gemma 3n E4B IT, 8B, score 1317.2 — on the efficiency frontier (best score at its size or smaller).
- Molmo2 8B, 9B, score 1328.0 — on the efficiency frontier (best score at its size or smaller).
- Gemma 3 12B IT, 12B, score 1341.7 — on the efficiency frontier (best score at its size or smaller).
- Mistral Small 3.2 24B Instruct 2506, 24B, score 1356.7 — on the efficiency frontier (best score at its size or smaller).
- Gemma 4 26B A4B IT, 26B, score 1437.9 — on the efficiency frontier (best score at its size or smaller).
- Gemma 4 31B IT, 31B, score 1451.1 — on the efficiency frontier (best score at its size or smaller).
- Hy3, 299B, score 1455.8 — on the efficiency frontier (best score at its size or smaller).
- GLM 5.3 Flash, 321B, score 1475.4 — on the efficiency frontier (best score at its size or smaller).
- GLM 5.3, 753B, score 1483.0 — on the efficiency frontier (best score at its size or smaller).
- Kimi K3, 2.8T, score 1484.8 — on the efficiency frontier (best score at its size or smaller).
LMArena Text: frequently asked questions
- What is the best open LLM on LMArena Text?
- Kimi K3 is the top open model on LMArena Text, scoring 1484.8. Among all models tested — including proprietary ones — it ranks #17. The top model overall is Claude Fable 5 (Anthropic) at 1505.7.
- What's the best LMArena Text model you can run on a 24 GB GPU?
- Gemma 4 31B IT is the highest-scoring open model that fits in 24 GB at 4-bit quantization (about 17 GB), scoring 1451.1 on LMArena Text.
- What's the best LMArena Text model you can run on a 12 GB GPU?
- Gemma 3 12B IT is the highest-scoring open model that fits in 12 GB at 4-bit quantization (about 7 GB), scoring 1341.7 on LMArena Text.
- Can open models match proprietary models on LMArena Text?
- Not quite on LMArena Text: the strongest proprietary model (Claude Fable 5) scores 1505.7, ahead of the best open model (Kimi K3) at 1484.8 — but you can run the open one yourself.
Scores aggregated from lmarena. llmrun does not run this benchmark — see the source for methodology, or the about benchmarks for what it measures.