Skip to content

/ LLM benchmarks / Language

Language benchmarks

Chat preference rankings (Text Arena Elo) and language comprehension (LiveBench).

DataText Arena2 Oct 2026LiveBenchrelease 2026-06-25fetched

Text Arena leaderboard, 413 models
#ModelScore
1
1525Elo ±9
2
1505Elo ±3
3
1504Elo ±4
4
1504Elo ±9
5
1501Elo ±4
6
1501Elo ±6
7
1497Elo ±3
8
1495Elo ±5
9
1494Elo ±6
10
1494Elo ±4
11
1494Elo ±9
12
1491Elo ±4
13
1490Elo ±4
14
1489Elo ±5
15
1489Elo ±6
16
Kimi K3 MaxMoonshot AI
1488Elo ±5
17
1488Elo ±5
18
1487Elo ±3
19
1485Elo ±4
20
1484Elo ±4
21
1483Elo ±11
22
1483Elo ±4
23
1482Elo ±5
24
1482Elo ±4
25
1481Elo ±4
26
1480Elo ±9
27
1478Elo ±6
28
1477Elo ±4
29
1477Elo ±7
30
GPT-5.5OpenAI
1477Elo ±4
31
1476Elo ±4
32
1476Elo ±4
33
1476Elo ±4
34
1475Elo ±4
35
1475Elo ±10
36
1475Elo ±5
37
1475Elo ±4
38
1474Elo ±7
39
1474Elo ±4
40
1473Elo ±5
41
1473Elo ±5
42
1473Elo ±4
43
1472Elo ±4
44
1472Elo ±4
45
1471Elo ±10
46
1471Elo ±4
47
1470Elo ±3
48
1468Elo ±5
49
1468Elo ±4
50
Grok 4.5SpaceXAI
1466Elo ±4
51
1465Elo ±4
52
1465Elo ±3
53
GPT-5.4OpenAI
1465Elo ±4
54
1465Elo ±5
55
1465Elo ±4
56
1464Elo ±7
57
1462Elo ±4
58
Kimi K2.6Moonshot AI
1461Elo ±4
59
1460Elo ±8
60
Grok 4.1SpaceXAI
1459Elo ±3
61
1458Elo ±3
62
1458Elo ±4
63
GLM-5Z.AI
1458Elo ±4
64
1457Elo ±7
65
1457Elo ±3
66
1456Elo ±3
67
Hy3Tencent
1456Elo ±7
68
1456Elo ±4
69
1456Elo ±4
70
1455Elo ±3
71
1455Elo ±4
72
1455Elo ±4
73
1454Elo ±5
74
1454Elo ±11
75
1453Elo ±4
76
1453Elo ±7
77
1452Elo ±8
78
1450Elo ±3
79
1450Elo ±4
80
1450Elo ±3
81
1449Elo ±7
82
1448Elo ±3
83
1448Elo ±5
84
1447Elo ±4
85
1446Elo ±4
86
1446Elo ±2
87
1445Elo ±6
88
1443Elo ±4
89
1443Elo ±3
90
1443Elo ±7
91
1442Elo ±8
92
1442Elo ±3
93
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
1438Elo ±4
99
1438Elo ±5
100
1437Elo ±7
101
1437Elo ±4
102
1437Elo ±5
103
1436Elo ±4
104
GPT-5.2OpenAI
1436Elo ±3
105
1435Elo ±5
106
1434Elo ±5
107
MiMo V2.5Xiaomi
1434Elo ±4
108
1433Elo ±6
109
1433Elo ±4
110
1432Elo ±4
111
1431Elo ±6
112
1431Elo ±7
113
1430Elo ±3
114
1430Elo ±3
115
1427Elo ±6
116
1427Elo ±4
117
1427Elo ±10
118
1426Elo ±7
119
1426Elo ±4
120
1425Elo ±10
121
1425Elo ±7
122
1425Elo ±3
123
1424Elo ±4
124
1423Elo ±7
125
1423Elo ±4
126
1422Elo ±3
127
1422Elo ±6
128
1422Elo ±6
129
1419Elo ±8
130
1419Elo ±9
131
1419Elo ±7
132
1418Elo ±5
133
1417Elo ±10
134
1417Elo ±6
135
1416Elo ±4
136
1416Elo ±7
137
1415Elo ±3
138
1415Elo ±10
139
1415Elo ±4
140
1414Elo ±4
141
1414Elo ±3
142
1414Elo ±2
143
1413Elo ±7
144
1413Elo ±10
145
1411Elo ±4
146
1411Elo ±5
147
Grok 4 0709SpaceXAI
1411Elo ±4
148
1409Elo ±2
149
1409Elo ±7
150
1409Elo ±4
151
1408Elo ±3
152
Inkling SmallThinking Machines
1405Elo ±5
153
1405Elo ±5
154
1403Elo ±4
155
1403Elo ±5
156
1402Elo ±4
157
1402Elo ±4
158
1401Elo ±4
159
1401Elo ±6
160
1400Elo ±7
161
1399Elo ±5
162
DeepSeek R1DeepSeek
1398Elo ±5
163
1396Elo ±4
164
1396Elo ±4
165
1395Elo ±7
166
1395Elo ±12
167
1394Elo ±9
168
1394Elo ±4
169
1393Elo ±4
170
1392Elo ±3
171
1391Elo ±4
172
1391Elo ±4
173
1391Elo ±4
174
1390Elo ±5
175
1389Elo ±5
176
1388Elo ±4
177
1388Elo ±5
178
1387Elo ±5
179
1387Elo ±9
180
1387Elo ±6
181
Solar Pro4Upstage
1386Elo ±6
182
1385Elo ±5
183
1383Elo ±4
184
1383Elo ±7
185
1383Elo ±5
186
1379Elo ±3
187
1379Elo ±11
188
1379Elo ±4
189
1375Elo ±5
190
1375Elo ±5
191
1374Elo ±3
192
1374Elo ±4
193
1373Elo ±4
194
1373Elo ±4
195
1369Elo ±6
196
1367Elo ±5
197
1365Elo ±4
198
1365Elo ±6
199
1364Elo ±4
200
MiniMax M1MiniMax
1364Elo ±4
201
1364Elo ±5
202
1363Elo ±5
203
1360Elo ±7
204
1360Elo ±4
205
DeepSeek V3DeepSeek
1358Elo ±5
206
1358Elo ±5
207
1356Elo ±5
208
1356Elo ±8
209
1354Elo ±3
210
1354Elo ±4
211
1352Elo ±8
212
1352Elo ±4
213
1351Elo ±3
214
1350Elo ±6
215
Step 3Stepfun
1349Elo ±8
216
1349Elo ±12
217
1348Elo ±6
218
o3 MiniOpenAI
1348Elo ±4
219
1348Elo ±12
220
1347Elo ±11
221
Qwen 3 32BAlibaba
1347Elo ±9
222
1346Elo ±3
223
1346Elo ±8
224
1344Elo ±7
225
MiniMax M2MiniMax
1343Elo ±8
226
Mercury 2Inception AI
1343Elo ±11
227
1343Elo ±3
228
1343Elo ±10
229
1343Elo ±8
230
1342Elo ±10
231
1341Elo ±12
232
1340Elo ±10
233
1338Elo ±7
234
o1 MiniOpenAI
1337Elo ±4
235
1336Elo ±5
236
Qwq 32BAlibaba
1336Elo ±4
237
1336Elo ±4
238
1336Elo ±4
239
1335Elo ±4
240
1335Elo ±6
241
1334Elo ±9
242
1334Elo ±4
243
1329Elo ±6
244
1328Elo ±5
245
1328Elo ±12
246
1327Elo ±4
247
1326Elo ±5
248
1326Elo ±10
249
1325Elo ±3
250
1324Elo ±4
251
1324Elo ±4
252
1323Elo ±8
253
1323Elo ±7
254
1322Elo ±8
255
1322Elo ±3
256
1322Elo ±21
257
1322Elo ±5
258
1320Elo ±7
259
1319Elo ±5
260
1318Elo ±6
261
1318Elo ±4
262
1318Elo ±4
263
1318Elo ±6
264
1317Elo ±5
265
1315Elo ±6
266
1314Elo ±4
267
1314Elo ±5
268
1314Elo ±5
269
1313Elo ±4
270
1313Elo ±4
271
1311Elo ±10
272
1309Elo ±4
273
1308Elo ±4
274
1307Elo ±5
275
1307Elo ±6
276
1306Elo ±8
277
1306Elo ±4
278
MercuryInception AI
1305Elo ±14
279
1305Elo ±6
280
1304Elo ±10
281
1303Elo ±9
282
1303Elo ±5
283
1303Elo ±4
284
1299Elo ±8
285
1294Elo ±9
286
1293Elo ±4
287
1290Elo ±5
288
1290Elo ±12
289
1289Elo ±3
290
1289Elo ±7
291
1288Elo ±11
292
1288Elo ±7
293
1288Elo ±5
294
1287Elo ±8
295
1287Elo ±10
296
1286Elo ±5
297
1286Elo ±7
298
1286Elo ±10
299
1281Elo ±4
300
1280Elo ±7
301
1277Elo ±5
302
1277Elo ±4
303
1276Elo ±7
304
1276Elo ±4
305
1274Elo ±6
306
1273Elo ±7
307
1272Elo ±7
308
1271Elo ±8
309
1267Elo ±5
310
1267Elo ±4
311
1265Elo ±6
312
1262Elo ±5
313
1262Elo ±4
314
1262Elo ±4
315
1260Elo ±5
316
1259Elo ±4
317
Phi 4Microsoft
1256Elo ±5
318
1252Elo ±11
319
1250Elo ±7
320
1242Elo ±5
321
1241Elo ±5
322
1240Elo ±7
323
1238Elo ±9
324
1237Elo ±7
325
1234Elo ±6
326
1233Elo ±7
327
1233Elo ±5
328
1233Elo ±12
329
1230Elo ±5
330
Command RCohere
1227Elo ±5
331
1227Elo ±6
332
1226Elo ±5
333
1224Elo ±4
334
1224Elo ±12
335
1223Elo ±7
336
1223Elo ±5
337
1220Elo ±11
338
1213Elo ±11
339
1213Elo ±5
340
1211Elo ±4
341
1208Elo ±11
342
1204Elo ±9
343
1204Elo ±6
344
1200Elo ±4
345
1198Elo ±5
346
1197Elo ±4
347
1195Elo ±6
348
1191Elo ±7
349
1191Elo ±7
350
1185Elo ±11
351
Wizardlm 70BMicrosoft
1185Elo ±9
352
1184Elo ±7
353
1183Elo ±9
354
1183Elo ±10
355
1183Elo ±8
356
1183Elo ±6
357
1180Elo ±6
358
1179Elo ±11
359
1178Elo ±10
360
1176Elo ±10
361
Vicuna 33BUnknown
1173Elo ±6
362
1171Elo ±6
363
1171Elo ±7
364
1171Elo ±5
365
1167Elo ±8
366
1167Elo ±8
367
1164Elo ±12
368
1157Elo ±8
369
1155Elo ±13
370
1154Elo ±11
371
1152Elo ±13
372
1152Elo ±15
373
1151Elo ±12
374
Wizardlm 13BMicrosoft
1149Elo ±9
375
1149Elo ±7
376
1148Elo ±17
377
1144Elo ±10
378
1143Elo ±6
379
Vicuna 13BUnknown
1142Elo ±7
380
1141Elo ±7
381
1139Elo ±11
382
Palm 2Google
1139Elo ±9
383
1138Elo ±9
384
1137Elo ±9
385
1131Elo ±9
386
1130Elo ±7
387
1128Elo ±6
388
1127Elo ±12
389
1127Elo ±16
390
1122Elo ±11
391
1119Elo ±18
392
1117Elo ±8
393
Vicuna 7BUnknown
1115Elo ±9
394
1115Elo ±14
395
1111Elo ±8
396
1110Elo ±9
397
1108Elo ±7
398
1094Elo ±11
399
1091Elo ±9
400
1074Elo ±11
401
Koala 13BUnknown
1071Elo ±10
402
Alpaca 13BUnknown
1070Elo ±11
403
1068Elo ±15
404
1063Elo ±12
405
1057Elo ±12
406
1042Elo ±11
407
1025Elo ±14
408
1023Elo ±11
409
996Elo ±12
410
992Elo ±12
411
982Elo ±13
412
975Elo ±16
413
953Elo ±13

Need a model picked for your use case?

Benchmarks rank models on other people's tasks. For a business workflow I test the shortlist on your own data, costs and privacy requirements, and recommend one.

AI consulting