/ LLM-Benchmarks / Sprache
Beste KI für Texte: welche Antworten Menschen bevorzugen
Blinde Abstimmungen, welche von zwei Antworten Menschen vorziehen (Text Arena Elo), dazu Sprachverständnis (LiveBench).
DatenText Arena02.10.2026LiveBenchRelease 2026-06-25abgerufen
Text Arena
Originalquelle ansehen| # | Modell | Wertung |
|---|---|---|
| 1 | Gemini 4 Argon HighGoogle | 1525Elo ±9 |
| 2 | Claude Opus 4.6 HighAnthropic | 1505Elo ±3 |
| 3 | Claude Fable 5 HighAnthropic | 1504Elo ±4 |
| 4 | Claude Opus 5.5 HighAnthropic | 1504Elo ±9 |
| 5 | Claude Opus 4.7 HighAnthropic | 1501Elo ±4 |
| 6 | Claude Fable 5.1 MaxAnthropic | 1501Elo ±6 |
| 7 | Claude Opus 4.6Anthropic | 1497Elo ±3 |
| 8 | Gemini 3.8 Flash HighGoogle | 1495Elo ±5 |
| 9 | 1494Elo ±6 | |
| 10 | Claude Opus 4.7Anthropic | 1494Elo ±4 |
| 11 | 1494Elo ±9 | |
| 12 | Muse Spark 1.1Meta | 1491Elo ±4 |
| 13 | Claude Opus 5 HighAnthropic | 1490Elo ±4 |
| 14 | Claude Opus 5 MaxAnthropic | 1489Elo ±5 |
| 15 | Muse SparkMeta | 1489Elo ±6 |
| 16 | Kimi K3 MaxMoonshot AI | 1488Elo ±5 |
| 17 | Gemini 3.7 Flash HighGoogle | 1488Elo ±5 |
| 18 | Gemini 3.1 Pro PreviewGoogle | 1487Elo ±3 |
| 19 | Gemini 3 ProGoogle | 1485Elo ±4 |
| 20 | GPT-5.6 Sol xHighOpenAI | 1484Elo ±4 |
| 21 | GPT-6.1 Sol MaxOpenAI | 1483Elo ±11 |
| 22 | Gemini 3.6 Flash HighGoogle | 1483Elo ±4 |
| 23 | Qwen 3.8 MaxAlibaba | 1482Elo ±5 |
| 24 | Claude Opus 4.8 HighAnthropic | 1482Elo ±4 |
| 25 | GPT-5.5 HighOpenAI | 1481Elo ±4 |
| 26 | MiMo V2.6 ProXiaomi | 1480Elo ±9 |
| 27 | GLM-5.3 MaxZ.AI | 1478Elo ±6 |
| 28 | Gemini 3.5 Flash HighGoogle | 1477Elo ±4 |
| 29 | GPT-6 Astra MaxOpenAI | 1477Elo ±7 |
| 30 | GPT-5.5OpenAI | 1477Elo ±4 |
| 31 | Gemini 3.5 Flash MediumGoogle | 1476Elo ±4 |
| 32 | GLM-5.2 MaxZ.AI | 1476Elo ±4 |
| 33 | 1476Elo ±4 | |
| 34 | GPT-5.4 HighOpenAI | 1475Elo ±4 |
| 35 | Qwen 3.7 Max PreviewAlibaba | 1475Elo ±10 |
| 36 | Grok 4.20 Beta1SpaceXAI | 1475Elo ±5 |
| 37 | Claude Opus 4.8Anthropic | 1475Elo ±4 |
| 38 | DeepSeek V4.1 Flash MaxDeepSeek | 1474Elo ±7 |
| 39 | Claude Opus 4.5 20251101 High 32KAnthropic | 1474Elo ±4 |
| 40 | GPT-5.5 InstantOpenAI | 1473Elo ±5 |
| 41 | GLM-5.3 FlashZ.AI | 1473Elo ±5 |
| 42 | Gemini 3 FlashGoogle | 1473Elo ±4 |
| 43 | Claude Sonnet 4.6Anthropic | 1472Elo ±4 |
| 44 | Grok 4.20 Beta 0309 ReasoningSpaceXAI | 1472Elo ±4 |
| 45 | Claude Sonnet 5.5 xHighAnthropic | 1471Elo ±10 |
| 46 | Grok 4.20 Multi Agent Beta 0309SpaceXAI | 1471Elo ±4 |
| 47 | Claude Opus 4.5 20251101Anthropic | 1470Elo ±3 |
| 48 | ERNIE 5.1Baidu | 1468Elo ±5 |
| 49 | MiMo V2.5 ProXiaomi | 1468Elo ±4 |
| 50 | Grok 4.5SpaceXAI | 1466Elo ±4 |
| 51 | GPT-5.6 Terra xHighOpenAI | 1465Elo ±4 |
| 52 | Grok 4.1 ThinkingSpaceXAI | 1465Elo ±3 |
| 53 | GPT-5.4OpenAI | 1465Elo ±4 |
| 54 | Qwen 3.5 Max PreviewAlibaba | 1465Elo ±5 |
| 55 | GLM-5.1Z.AI | 1465Elo ±4 |
| 56 | DeepSeek V4 Pro High 20260813DeepSeek | 1464Elo ±7 |
| 57 | Claude Sonnet 5 HighAnthropic | 1462Elo ±4 |
| 58 | Kimi K2.6Moonshot AI | 1461Elo ±4 |
| 59 | Qwen 3.6 Max PreviewAlibaba | 1460Elo ±8 |
| 60 | Grok 4.1SpaceXAI | 1459Elo ±3 |
| 61 | 1458Elo ±3 | |
| 62 | DeepSeek V4 ProDeepSeek | 1458Elo ±4 |
| 63 | GLM-5Z.AI | 1458Elo ±4 |
| 64 | GPT-6 Sol MaxOpenAI | 1457Elo ±7 |
| 65 | Claude Sonnet 4.5 20250929 High 32KAnthropic | 1457Elo ±3 |
| 66 | Dola Seed 2.0 ProBytedance | 1456Elo ±3 |
| 67 | Hy3Tencent | 1456Elo ±7 |
| 68 | Qwen 3.7 PlusAlibaba | 1456Elo ±4 |
| 69 | GPT-5.1 HighOpenAI | 1456Elo ±4 |
| 70 | Claude Sonnet 4.5 20250929Anthropic | 1455Elo ±3 |
| 71 | Gemini 3.5 Flash LiteGoogle | 1455Elo ±4 |
| 72 | DeepSeek V4 Pro High PreviewDeepSeek | 1455Elo ±4 |
| 73 | Grok 4.6 HighSpaceXAI | 1454Elo ±5 |
| 74 | Step 5 PreviewStepfun | 1454Elo ±11 |
| 75 | GPT-5.6 Luna xHighOpenAI | 1453Elo ±4 |
| 76 | Gemma 4 31BGoogle | 1453Elo ±7 |
| 77 | MiMo V2.6 FlashXiaomi | 1452Elo ±8 |
| 78 | Kimi K2.5 ThinkingMoonshot AI | 1450Elo ±3 |
| 79 | GPT-5.3 Chat LatestOpenAI | 1450Elo ±4 |
| 80 | 1450Elo ±3 | |
| 81 | 1449Elo ±7 | |
| 82 | Claude Opus 4.1 20250805Anthropic | 1448Elo ±3 |
| 83 | MiMo V2 ProXiaomi | 1448Elo ±5 |
| 84 | GPT-5.4 Mini HighOpenAI | 1447Elo ±4 |
| 85 | ERNIE 5.0 0110Baidu | 1446Elo ±4 |
| 86 | Gemini 2.5 ProGoogle | 1446Elo ±2 |
| 87 | 1445Elo ±6 | |
| 88 | Qwen 3.6 PlusAlibaba | 1443Elo ±4 |
| 89 | 1443Elo ±3 | |
| 90 | GPT-6 Luna MaxOpenAI | 1443Elo ±7 |
| 91 | Grok 4.7 xHighSpaceXAI | 1442Elo ±8 |
| 92 | Qwen 3.5 397B A17BAlibaba | 1442Elo ±3 |
| 93 | GLM-4.7Z.AI | 1441Elo ±6 |
| 94 | Grok 4.3SpaceXAI | 1441Elo ±4 |
| 95 | InklingThinking Machines | 1441Elo ±5 |
| 96 | MiniMax M3MiniMax | 1440Elo ±4 |
| 97 | GPT-5.1OpenAI | 1439Elo ±4 |
| 98 | DeepSeek V4 Flash High PreviewDeepSeek | 1438Elo ±4 |
| 99 | Qwen 3.8 27BAlibaba | 1438Elo ±5 |
| 100 | Gemma 4 26B A4BGoogle | 1437Elo ±7 |
| 101 | GPT-5.2 HighOpenAI | 1437Elo ±4 |
| 102 | Longcat Flash Chat 2602 ExpMeituan | 1437Elo ±5 |
| 103 | DeepSeek V4 FlashDeepSeek | 1436Elo ±4 |
| 104 | GPT-5.2OpenAI | 1436Elo ±3 |
| 105 | GPT-5 HighOpenAI | 1435Elo ±5 |
| 106 | Qwen 3 Max PreviewAlibaba | 1434Elo ±5 |
| 107 | MiMo V2.5Xiaomi | 1434Elo ±4 |
| 108 | GLM-5v TurboZ.AI | 1433Elo ±6 |
| 109 | 1433Elo ±4 | |
| 110 | o3 2025-04-16OpenAI | 1432Elo ±4 |
| 111 | MiMo V2 OmniXiaomi | 1431Elo ±6 |
| 112 | Kimi K2.5 InstantMoonshot AI | 1431Elo ±7 |
| 113 | Grok 4.1 Fast ReasoningSpaceXAI | 1430Elo ±3 |
| 114 | Kimi K2 Thinking TurboMoonshot AI | 1430Elo ±3 |
| 115 | Mistral Medium 3.5Mistral | 1427Elo ±6 |
| 116 | GPT-5 ChatOpenAI | 1427Elo ±4 |
| 117 | 1427Elo ±10 | |
| 118 | 1426Elo ±7 | |
| 119 | Claude Opus 4 20250514 Thinking 16KAnthropic | 1426Elo ±4 |
| 120 | Muse GlimmerMeta | 1425Elo ±10 |
| 121 | DeepSeek V3.2 Exp ThinkingDeepSeek | 1425Elo ±7 |
| 122 | DeepSeek V3.2DeepSeek | 1425Elo ±3 |
| 123 | GLM-4.6Z.AI | 1424Elo ±4 |
| 124 | Qwen 3 Max 2025-09-23Alibaba | 1423Elo ±7 |
| 125 | DeepSeek V3.2 ThinkingDeepSeek | 1423Elo ±4 |
| 126 | 1422Elo ±3 | |
| 127 | DeepSeek V3.2 ExpDeepSeek | 1422Elo ±6 |
| 128 | DeepSeek R1 0528DeepSeek | 1422Elo ±6 |
| 129 | Grok 4 Fast ChatSpaceXAI | 1419Elo ±8 |
| 130 | 1419Elo ±9 | |
| 131 | Kimi K2 0905 PreviewMoonshot AI | 1419Elo ±7 |
| 132 | Kimi K2 0711 PreviewMoonshot AI | 1418Elo ±5 |
| 133 | DeepSeek V3.1 Terminus ThinkingDeepSeek | 1417Elo ±10 |
| 134 | DeepSeek V3.1DeepSeek | 1417Elo ±6 |
| 135 | Qwen 3.5 122B A10BAlibaba | 1416Elo ±4 |
| 136 | DeepSeek V3.1 ThinkingDeepSeek | 1416Elo ±7 |
| 137 | MiniMax M2.7MiniMax | 1415Elo ±3 |
| 138 | DeepSeek V3.1 TerminusDeepSeek | 1415Elo ±10 |
| 139 | GPT-4.1 2025-04-14OpenAI | 1415Elo ±4 |
| 140 | Claude Opus 4 20250514Anthropic | 1414Elo ±4 |
| 141 | Mistral Large 3Mistral | 1414Elo ±3 |
| 142 | Claude Haiku 4.5 20251001Anthropic | 1414Elo ±2 |
| 143 | Qwen 3 Vl 235B A22B InstructAlibaba | 1413Elo ±7 |
| 144 | 1413Elo ±10 | |
| 145 | Grok 3 Preview 02 24SpaceXAI | 1411Elo ±4 |
| 146 | GLM-4.5Z.AI | 1411Elo ±5 |
| 147 | Grok 4 0709SpaceXAI | 1411Elo ±4 |
| 148 | Gemini 2.5 FlashGoogle | 1409Elo ±2 |
| 149 | Hunyuan Hy3 PreviewTencent | 1409Elo ±7 |
| 150 | Qwen 3.5 27BAlibaba | 1409Elo ±4 |
| 151 | Mistral Medium 2508Mistral | 1408Elo ±3 |
| 152 | Inkling SmallThinking Machines | 1405Elo ±5 |
| 153 | Grok 4 Fast ReasoningSpaceXAI | 1405Elo ±5 |
| 154 | 1403Elo ±4 | |
| 155 | Qwen 3 235B A22B No ThinkingAlibaba | 1403Elo ±5 |
| 156 | o1 2024-12-17OpenAI | 1402Elo ±4 |
| 157 | 1402Elo ±4 | |
| 158 | GPT-5.4 Nano HighOpenAI | 1401Elo ±4 |
| 159 | Longcat Flash ChatMeituan | 1401Elo ±6 |
| 160 | 1400Elo ±7 | |
| 161 | Qwen 3 Next 80B A3B InstructAlibaba | 1399Elo ±5 |
| 162 | DeepSeek R1DeepSeek | 1398Elo ±5 |
| 163 | Qwen 3.5 FlashAlibaba | 1396Elo ±4 |
| 164 | DeepSeek V3 0324DeepSeek | 1396Elo ±4 |
| 165 | Qwen 3 Vl 235B A22B ThinkingAlibaba | 1395Elo ±7 |
| 166 | Hunyuan Vision 1.5 ThinkingTencent | 1395Elo ±12 |
| 167 | 1394Elo ±9 | |
| 168 | Qwen 3.5 35B A3BAlibaba | 1394Elo ±4 |
| 169 | Step 3.5 FlashStepfun | 1393Elo ±4 |
| 170 | 1392Elo ±3 | |
| 171 | MiniMax M2.5MiniMax | 1391Elo ±4 |
| 172 | o4 Mini 2025-04-16OpenAI | 1391Elo ±4 |
| 173 | Claude Sonnet 4 20250514Anthropic | 1391Elo ±4 |
| 174 | GPT-5 Mini HighOpenAI | 1390Elo ±5 |
| 175 | o1 PreviewOpenAI | 1389Elo ±5 |
| 176 | 1388Elo ±4 | |
| 177 | Mistral Medium 2505Mistral | 1388Elo ±5 |
| 178 | 1387Elo ±5 | |
| 179 | Hunyuan T1 20250711Tencent | 1387Elo ±9 |
| 180 | MiMo V2 Flash (thinking)Xiaomi | 1387Elo ±6 |
| 181 | Solar Pro4Upstage | 1386Elo ±6 |
| 182 | MiniMax M2.1 PreviewMiniMax | 1385Elo ±5 |
| 183 | GPT-4.1 Mini 2025-04-14OpenAI | 1383Elo ±4 |
| 184 | Hunyuan Turbos 20250416Tencent | 1383Elo ±7 |
| 185 | Qwen 3 30B A3B Instruct 2507Alibaba | 1383Elo ±5 |
| 186 | 1379Elo ±3 | |
| 187 | GLM-4.6vZ.AI | 1379Elo ±11 |
| 188 | Trinity Large PreviewArcee AI | 1379Elo ±4 |
| 189 | Qwen 3 235B A22BAlibaba | 1375Elo ±5 |
| 190 | 1375Elo ±5 | |
| 191 | Claude 3.5 Sonnet 20241022Anthropic | 1374Elo ±3 |
| 192 | Qwen 2.5 MaxAlibaba | 1374Elo ±4 |
| 193 | GLM-4.5 AirZ.AI | 1373Elo ±4 |
| 194 | Claude 3.7 Sonnet 20250219Anthropic | 1373Elo ±4 |
| 195 | Qwen 3 Next 80B A3B ThinkingAlibaba | 1369Elo ±6 |
| 196 | Trinity Large ThinkingArcee AI | 1367Elo ±5 |
| 197 | Gemma 3 27B ItGoogle | 1365Elo ±4 |
| 198 | GLM-4.7 FlashZ.AI | 1365Elo ±6 |
| 199 | 1364Elo ±4 | |
| 200 | MiniMax M1MiniMax | 1364Elo ±4 |
| 201 | o3 Mini HighOpenAI | 1364Elo ±5 |
| 202 | Grok 3 Mini HighSpaceXAI | 1363Elo ±5 |
| 203 | 1360Elo ±7 | |
| 204 | Gemini 2.0 Flash 001Google | 1360Elo ±4 |
| 205 | DeepSeek V3DeepSeek | 1358Elo ±5 |
| 206 | Grok 3 Mini BetaSpaceXAI | 1358Elo ±5 |
| 207 | Mistral Small 2506Mistral | 1356Elo ±5 |
| 208 | Intellect 3Unknown | 1356Elo ±8 |
| 209 | Command A 03 2025Cohere | 1354Elo ±3 |
| 210 | 1354Elo ±4 | |
| 211 | GLM-4.5vZ.AI | 1352Elo ±8 |
| 212 | GPT-OSS 120BOpenAI | 1352Elo ±4 |
| 213 | Gemini 1.5 Pro 002Google | 1351Elo ±3 |
| 214 | 1350Elo ±6 | |
| 215 | Step 3Stepfun | 1349Elo ±8 |
| 216 | Hunyuan Turbos 20250226Tencent | 1349Elo ±12 |
| 217 | 1348Elo ±6 | |
| 218 | o3 MiniOpenAI | 1348Elo ±4 |
| 219 | 1348Elo ±12 | |
| 220 | 1347Elo ±11 | |
| 221 | Qwen 3 32BAlibaba | 1347Elo ±9 |
| 222 | GPT-4o 2024-05-13OpenAI | 1346Elo ±3 |
| 223 | Qwen Plus 0125Alibaba | 1346Elo ±8 |
| 224 | Ling Flash 2.0Ant Group | 1344Elo ±7 |
| 225 | MiniMax M2MiniMax | 1343Elo ±8 |
| 226 | Mercury 2Inception AI | 1343Elo ±11 |
| 227 | Claude 3.5 Sonnet 20240620Anthropic | 1343Elo ±3 |
| 228 | 1343Elo ±10 | |
| 229 | GLM-4 Plus 0111Z.AI | 1343Elo ±8 |
| 230 | Gemma 3 12B ItGoogle | 1342Elo ±10 |
| 231 | Hunyuan Turbo 0110Tencent | 1341Elo ±12 |
| 232 | 1340Elo ±10 | |
| 233 | GPT-5 Nano HighOpenAI | 1338Elo ±7 |
| 234 | o1 MiniOpenAI | 1337Elo ±4 |
| 235 | Gemini Advanced 0514Google | 1336Elo ±5 |
| 236 | Qwq 32BAlibaba | 1336Elo ±4 |
| 237 | Grok 2 2024-08-13SpaceXAI | 1336Elo ±4 |
| 238 | GPT-4o 2024-08-06OpenAI | 1336Elo ±4 |
| 239 | 1335Elo ±4 | |
| 240 | Nova 2 LiteAmazon | 1335Elo ±6 |
| 241 | Step 2 16K Exp 202412Stepfun | 1334Elo ±9 |
| 242 | 1334Elo ±4 | |
| 243 | 1329Elo ±6 | |
| 244 | Yi LightningUnknown | 1328Elo ±5 |
| 245 | 1328Elo ±12 | |
| 246 | 1327Elo ±4 | |
| 247 | Qwen 3 30B A3BAlibaba | 1326Elo ±5 |
| 248 | Hunyuan Large 2025-02-10Tencent | 1326Elo ±10 |
| 249 | Claude 3.5 Haiku 20241022Anthropic | 1325Elo ±3 |
| 250 | GPT-4 Turbo 2024-04-09OpenAI | 1324Elo ±4 |
| 251 | Gemini 1.5 Pro 001Google | 1324Elo ±4 |
| 252 | DeepSeek V2.5 1210DeepSeek | 1323Elo ±8 |
| 253 | Ring Flash 2.0Ant Group | 1323Elo ±7 |
| 254 | GPT-4.1 Nano 2025-04-14OpenAI | 1322Elo ±8 |
| 255 | Claude 3 Opus 20240229Anthropic | 1322Elo ±3 |
| 256 | Molmo 2 8BAi2 | 1322Elo ±21 |
| 257 | 1322Elo ±5 | |
| 258 | Step 1o Turbo 202506Stepfun | 1320Elo ±7 |
| 259 | GLM-4 PlusZ.AI | 1319Elo ±5 |
| 260 | Qwen Max 0919Alibaba | 1318Elo ±6 |
| 261 | 1318Elo ±4 | |
| 262 | GPT-4o Mini 2024-07-18OpenAI | 1318Elo ±4 |
| 263 | GPT-OSS 20BOpenAI | 1318Elo ±6 |
| 264 | Gemma 3n E4B ItGoogle | 1317Elo ±5 |
| 265 | Qwen 2.5 Plus 1127Alibaba | 1315Elo ±6 |
| 266 | Mistral Large 2407Mistral | 1314Elo ±4 |
| 267 | Athene V2 ChatUnknown | 1314Elo ±5 |
| 268 | 1314Elo ±5 | |
| 269 | GPT-4 0125 PreviewOpenAI | 1313Elo ±4 |
| 270 | GPT-4 1106 PreviewOpenAI | 1313Elo ±4 |
| 271 | Hunyuan Standard 2025-02-10Tencent | 1311Elo ±10 |
| 272 | Gemini 1.5 Flash 002Google | 1309Elo ±4 |
| 273 | Grok 2 Mini 2024-08-13SpaceXAI | 1308Elo ±4 |
| 274 | DeepSeek V2.5DeepSeek | 1307Elo ±5 |
| 275 | Athene 70B 0725Unknown | 1307Elo ±6 |
| 276 | 1306Elo ±8 | |
| 277 | Mistral Large 2411Mistral | 1306Elo ±4 |
| 278 | MercuryInception AI | 1305Elo ±14 |
| 279 | Magistral Medium 2506Mistral | 1305Elo ±6 |
| 280 | 1304Elo ±10 | |
| 281 | Gemma 3 4B ItGoogle | 1303Elo ±9 |
| 282 | 1303Elo ±5 | |
| 283 | Qwen 2.5 72B InstructAlibaba | 1303Elo ±4 |
| 284 | 1299Elo ±8 | |
| 285 | Hunyuan Large VisionTencent | 1294Elo ±9 |
| 286 | 1293Elo ±4 | |
| 287 | Amazon Nova Pro V1.0Amazon | 1290Elo ±5 |
| 288 | 1290Elo ±12 | |
| 289 | Gemma 2 27B ItGoogle | 1289Elo ±3 |
| 290 | Jamba 1.5 LargeUnknown | 1289Elo ±7 |
| 291 | 1288Elo ±11 | |
| 292 | Reka Core 20240904Unknown | 1288Elo ±7 |
| 293 | GPT-4 0314OpenAI | 1288Elo ±5 |
| 294 | 1287Elo ±8 | |
| 295 | 1287Elo ±10 | |
| 296 | Gemini 1.5 Flash 001Google | 1286Elo ±5 |
| 297 | 1286Elo ±7 | |
| 298 | 1286Elo ±10 | |
| 299 | Claude 3 Sonnet 20240229Anthropic | 1281Elo ±4 |
| 300 | Gemma 2 9B It SimpoGoogle | 1280Elo ±7 |
| 301 | Nemotron 4 340B InstructNVIDIA | 1277Elo ±5 |
| 302 | 1277Elo ±4 | |
| 303 | Command R Plus 08 2024Cohere | 1276Elo ±7 |
| 304 | GPT-4 0613OpenAI | 1276Elo ±4 |
| 305 | 1274Elo ±6 | |
| 306 | GLM-4 0520Z.AI | 1273Elo ±7 |
| 307 | Reka Flash 20240904Unknown | 1272Elo ±7 |
| 308 | Qwen 2.5 Coder 32B InstructAlibaba | 1271Elo ±8 |
| 309 | C4ai Aya Expanse 32BCohere | 1267Elo ±5 |
| 310 | Gemma 2 9B ItGoogle | 1267Elo ±4 |
| 311 | DeepSeek Coder V2DeepSeek | 1265Elo ±6 |
| 312 | Qwen 2 72B InstructAlibaba | 1262Elo ±5 |
| 313 | Command R PlusCohere | 1262Elo ±4 |
| 314 | Claude 3 Haiku 20240307Anthropic | 1262Elo ±4 |
| 315 | Amazon Nova Lite V1.0Amazon | 1260Elo ±5 |
| 316 | Gemini 1.5 Flash 8B 001Google | 1259Elo ±4 |
| 317 | Phi 4Microsoft | 1256Elo ±5 |
| 318 | 1252Elo ±11 | |
| 319 | Command R 08 2024Cohere | 1250Elo ±7 |
| 320 | Mistral Large 2402Mistral | 1242Elo ±5 |
| 321 | Amazon Nova Micro V1.0Amazon | 1241Elo ±5 |
| 322 | Jamba 1.5 MiniUnknown | 1240Elo ±7 |
| 323 | Ministral 8B 2410Mistral | 1238Elo ±9 |
| 324 | Gemini Pro Dev ApiGoogle | 1237Elo ±7 |
| 325 | Qwen 1.5 110B ChatAlibaba | 1234Elo ±6 |
| 326 | 1233Elo ±7 | |
| 327 | Qwen 1.5 72B ChatAlibaba | 1233Elo ±5 |
| 328 | Hunyuan Standard 256KTencent | 1233Elo ±12 |
| 329 | Mixtral 8x22b Instruct V0.1Mistral | 1230Elo ±5 |
| 330 | Command RCohere | 1227Elo ±5 |
| 331 | Reka Flash 21B 20240226Unknown | 1227Elo ±6 |
| 332 | GPT-3.5 Turbo 0125OpenAI | 1226Elo ±5 |
| 333 | 1224Elo ±4 | |
| 334 | Gemini ProGoogle | 1224Elo ±12 |
| 335 | C4ai Aya Expanse 8BCohere | 1223Elo ±7 |
| 336 | Mistral MediumMistral | 1223Elo ±5 |
| 337 | 1220Elo ±11 | |
| 338 | Zephyr Orpo 141B A35B V0.1Unknown | 1213Elo ±11 |
| 339 | Yi 1.5 34B ChatUnknown | 1213Elo ±5 |
| 340 | 1211Elo ±4 | |
| 341 | 1208Elo ±11 | |
| 342 | GPT-3.5 Turbo 1106OpenAI | 1204Elo ±9 |
| 343 | Qwen 1.5 32B ChatAlibaba | 1204Elo ±6 |
| 344 | Gemma 2 2B ItGoogle | 1200Elo ±4 |
| 345 | Phi 3 Medium 4K InstructMicrosoft | 1198Elo ±5 |
| 346 | Mixtral 8x7b Instruct V0.1Mistral | 1197Elo ±4 |
| 347 | Dbrx Instruct PreviewUnknown | 1195Elo ±6 |
| 348 | Qwen 1.5 14B ChatAlibaba | 1191Elo ±7 |
| 349 | Internlm2 5 20B ChatUnknown | 1191Elo ±7 |
| 350 | DeepSeek Llm 67B ChatDeepSeek | 1185Elo ±11 |
| 351 | Wizardlm 70BMicrosoft | 1185Elo ±9 |
| 352 | Yi 34B ChatUnknown | 1184Elo ±7 |
| 353 | 1183Elo ±9 | |
| 354 | Openchat 3.5Unknown | 1183Elo ±10 |
| 355 | Openchat 3.5 0106Unknown | 1183Elo ±8 |
| 356 | Gemma 1.1 7B ItGoogle | 1183Elo ±6 |
| 357 | Snowflake Arctic InstructUnknown | 1180Elo ±6 |
| 358 | 1179Elo ±11 | |
| 359 | Tulu 2 Dpo 70BUnknown | 1178Elo ±10 |
| 360 | Openhermes 2.5 Mistral 7BUnknown | 1176Elo ±10 |
| 361 | Vicuna 33BUnknown | 1173Elo ±6 |
| 362 | Phi 3 Small 8K InstructMicrosoft | 1171Elo ±6 |
| 363 | Starling Lm 7B BetaUnknown | 1171Elo ±7 |
| 364 | Llama 2 70B ChatMeta | 1171Elo ±5 |
| 365 | Starling Lm 7B AlphaUnknown | 1167Elo ±8 |
| 366 | 1167Elo ±8 | |
| 367 | 1164Elo ±12 | |
| 368 | 1157Elo ±8 | |
| 369 | Llama2 70B Steerlm ChatNVIDIA | 1155Elo ±13 |
| 370 | Qwq 32B PreviewAlibaba | 1154Elo ±11 |
| 371 | Solar 10.7B Instruct V1.0Unknown | 1152Elo ±13 |
| 372 | Dolphin 2.2.1 Mistral 7BUnknown | 1152Elo ±15 |
| 373 | Mpt 30B ChatUnknown | 1151Elo ±12 |
| 374 | Wizardlm 13BMicrosoft | 1149Elo ±9 |
| 375 | Mistral 7B Instruct V0.2Mistral | 1149Elo ±7 |
| 376 | Falcon 180B ChatUnknown | 1148Elo ±17 |
| 377 | Qwen 1.5 7B ChatAlibaba | 1144Elo ±10 |
| 378 | Phi 3 Mini 4K Instruct June 2024Microsoft | 1143Elo ±6 |
| 379 | Vicuna 13BUnknown | 1142Elo ±7 |
| 380 | Llama 2 13B ChatMeta | 1141Elo ±7 |
| 381 | Qwen 14B ChatAlibaba | 1139Elo ±11 |
| 382 | Palm 2Google | 1139Elo ±9 |
| 383 | Gemma 7B ItGoogle | 1138Elo ±9 |
| 384 | 1137Elo ±9 | |
| 385 | Zephyr 7B BetaUnknown | 1131Elo ±9 |
| 386 | Phi 3 Mini 128K InstructMicrosoft | 1130Elo ±7 |
| 387 | Phi 3 Mini 4K InstructMicrosoft | 1128Elo ±6 |
| 388 | Guanaco 33BUnknown | 1127Elo ±12 |
| 389 | Zephyr 7B AlphaUnknown | 1127Elo ±16 |
| 390 | Stripedhyena Nous 7BUnknown | 1122Elo ±11 |
| 391 | 1119Elo ±18 | |
| 392 | Gemma 1.1 2B ItGoogle | 1117Elo ±8 |
| 393 | Vicuna 7BUnknown | 1115Elo ±9 |
| 394 | Smollm2 1.7B InstructUnknown | 1115Elo ±14 |
| 395 | 1111Elo ±8 | |
| 396 | Mistral 7B InstructMistral | 1110Elo ±9 |
| 397 | Llama 2 7B ChatMeta | 1108Elo ±7 |
| 398 | Gemma 2B ItGoogle | 1094Elo ±11 |
| 399 | Qwen 1.5 4B ChatAlibaba | 1091Elo ±9 |
| 400 | 1074Elo ±11 | |
| 401 | Koala 13BUnknown | 1071Elo ±10 |
| 402 | Alpaca 13BUnknown | 1070Elo ±11 |
| 403 | Gpt4all 13B SnoozyOpenAI | 1068Elo ±15 |
| 404 | Mpt 7B ChatUnknown | 1063Elo ±12 |
| 405 | Chatglm3 6BZ.AI | 1057Elo ±12 |
| 406 | RWKV 4 Raven 14BUnknown | 1042Elo ±11 |
| 407 | Chatglm2 6BZ.AI | 1025Elo ±14 |
| 408 | Oasst Pythia 12BUnknown | 1023Elo ±11 |
| 409 | Chatglm 6BZ.AI | 996Elo ±12 |
| 410 | Fastchat T5 3BUnknown | 992Elo ±12 |
| 411 | Dolly V2 12BUnknown | 982Elo ±13 |
| 412 | Llama 13BMeta | 975Elo ±16 |
| 413 | Stablelm Tuned Alpha 7BUnknown | 953Elo ±13 |
LiveBench Language · CC BY-SA 4.0
Originalquelle ansehen| # | Modell | Wertung | Kosten pro gelöster Aufgabe |
|---|---|---|---|
| 1 | Claude Fable 5 Max EffortAnthropic | 90.7% | $1.44 |
| 2 | GPT-6.1 Sol Max EffortOpenAI | 90.1% | $0.14 |
| 3 | Claude Fable 5.1 Max EffortAnthropic | 89.5% | $1.21 |
| 4 | GPT-6 Astra Max EffortOpenAI | 89.4% | $0.74 |
| 5 | Claude 5 Opus Thinking Max EffortAnthropic | 88.7% | $0.70 |
| 6 | Gemini 3.8 Flash HighGoogle | 87.8% | $0.31 |
| 7 | GPT-5.6 Sol Max EffortOpenAI | 87.7% | $0.52 |
| 8 | GPT-5.5 Thinking xHigh EffortOpenAI | 87.4% | $0.43 |
| 9 | Claude 5.5 Opus Thinking Max EffortAnthropic | 86.3% | $0.80 |
| 10 | Kimi K3Moonshot AI · Open Weights | 85.5% | $0.35 |
| 11 | Gemini 3.7 Flash HighGoogle | 85.5% | $0.16 |
| 12 | Gemini 3.1 Pro Preview HighGoogle | 85.4% | $0.29 |
| 13 | GPT-6 Sol Max EffortOpenAI | 85.3% | $0.27 |
| 14 | Gemini 3.5 Flash HighGoogle | 84.6% | $0.25 |
| 15 | Gemini 3.6 Flash HighGoogle | 83.9% | $0.23 |
| 16 | Grok 4.6 xHighxAI | 83.7% | $0.21 |
| 17 | Claude Sonnet 5.5 xHigh EffortAnthropic | 83.4% | $0.14 |
| 18 | Claude 4.6 Opus Thinking High EffortAnthropic | 83.3% | $0.40 |
| 19 | GPT-5.6 Terra Max EffortOpenAI | 82.9% | $0.35 |
| 20 | Muse Spark 1.3 xHigh EffortMeta | 82.8% | $0.22 |
| 21 | Grok 4.5xAI | 82.8% | $0.13 |
| 22 | GPT-5.4 Thinking xHigh EffortOpenAI | 82.6% | $0.39 |
| 23 | DeepSeek V4 Pro 0813DeepSeek · Open Weights | 82.1% | $0.044 |
| 24 | Claude 4.5 Opus Thinking High EffortAnthropic | 81.3% | $0.61 |
| 25 | DeepSeek V4.1 Flash Max EffortDeepSeek · Open Weights | 81.2% | $0.029 |
| 26 | DeepSeek V4 Flash Vision ExpDeepSeek · Open Weights | 80.4% | $0.051 |
| 27 | Grok 4.7 xHighxAI | 80.1% | $0.72 |
| 28 | GLM-5.3Z.AI · Open Weights | 79.9% | $0.45 |
| 29 | GPT-5.2 HighOpenAI | 79.8% | $0.23 |
| 30 | Qwen 3.8 MaxAlibaba · Open Weights | 79.7% | $0.28 |
| 31 | Claude 4.8 Opus Thinking Max EffortAnthropic | 79.7% | $0.98 |
| 32 | Qwen 3.7 MaxAlibaba | 79.7% | $0.18 |
| 33 | DeepSeek V4 Flash 0731DeepSeek · Open Weights | 79.2% | $0.060 |
| 34 | Muse Spark 1.2 xHigh EffortMeta | 78.6% | $0.38 |
| 35 | Claude 4.7 Opus Thinking xHigh EffortAnthropic | 77.9% | $0.53 |
| 36 | Kimi K2.7 CodeMoonshot AI · Open Weights | 77.9% | $0.10 |
| 37 | GLM-5.3 FlashZ.AI · Open Weights | 77.3% | $0.031 |
| 38 | Minimax M3MiniMax | 76.8% | $0.060 |
| 39 | GLM-5.2Z.AI · Open Weights | 76.2% | $0.23 |
| 40 | Claude 4.6 Sonnet Thinking Medium EffortAnthropic | 76.1% | $0.31 |
| 41 | Kimi K2.6 ThinkingMoonshot AI · Open Weights | 75.1% | $0.17 |
| 42 | Claude Sonnet 5 xHigh EffortAnthropic | 75.0% | $0.51 |
| 43 | Qwen 3.6 PlusAlibaba | 75.0% | $0.23 |
| 44 | Qwen 3.8 Flash NextAlibaba · Open Weights | 74.6% | $0.042 |
| 45 | Muse Spark 1.1 xHigh EffortMeta | 74.3% | $0.20 |
| 46 | Qwen3.8 27BAlibaba · Open Weights | 74.3% | $0.094 |
| 47 | GPT-6 Luna Max EffortOpenAI | 73.8% | $0.026 |
| 48 | GPT-5.2 CodexOpenAI | 73.7% | $0.19 |
| 49 | Grok 4.3xAI | 73.6% | $0.061 |
| 50 | Inkling xHigh EffortThinking Machines · Open Weights | 73.5% | $0.31 |
| 51 | GPT-5.6 Luna Max EffortOpenAI | 72.6% | $0.17 |
| 52 | Grok Build 0.1xAI | 72.5% | $0.024 |
| 53 | Gemini 3.5 Flash-Lite HighGoogle | 71.8% | $0.069 |
| 54 | GPT-5.4 Mini xHighOpenAI | 71.0% | $0.33 |
| 55 | Nemotron 3 Ultra 550B A55BNVIDIA · Open Weights | 70.8% | $0.37 |
| 56 | Ox Alpha MaxStealth | 66.1% | – |
| 57 | Qwen 3.6 27BAlibaba · Open Weights | 63.3% | $0.20 |
| 58 | GPT-5.4 Nano xHighOpenAI | 62.5% | $0.091 |
Sie brauchen ein Modell für Ihren konkreten Fall?
Benchmarks bewerten Modelle anhand fremder Aufgaben. Für einen Geschäftsprozess teste ich die engere Wahl mit Ihren eigenen Daten, Kosten und Datenschutzvorgaben und empfehle eines.