LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

LMArena Text: hard prompts in English leaderboard

Gemini 4 Argon leads
1548
models measured
156
lead over #2
+12 Elo
median
1431

Leader on LMArena Text: hard prompts in English · read October 21

1 Gemini 4 Argon1548

On October 21, 2026, 156 of our catalog's models had a LMArena Text: hard prompts in English result, and Google Gemini 4 Argon led with 1548.

Hard prompts written in English: specific, complex, needing problem solving. 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

156 models with a result

Original board: LMArena (Hugging Face dataset)
LMArena Text: hard prompts in English results for models in the catalog
# Model Result VotesInterval (±) Evidence Coding index Output $/Mtok
1 Gemini 4 ArgonGoogle 1548 1,208±17.3 80.5 $10.00
2 Claude Opus 5.5Anthropic 1535 2,275±13.1 92.2 $20.00
3 Claude Opus 4.6Anthropic 1530 23,533±5.2 66.9 $25.00
4 Claude Opus 5Anthropic 1523 8,691±7.3 81.8 $25.00
5 Claude Fable 5.1Anthropic 1522 3,681±10.3 89.1 $50.00
6 MiMo-V2.6-ProXiaomi 1520 940±18.8 72.6 $0.87
7 Claude Fable 5Anthropic 1509 12,023±6.3 82.1 $50.00
8 Claude Opus 4.7Anthropic 1508 20,828±5.6 51.1 $25.00
9 Muse WonderMeta 1506 1,282±16.7 64.5 $20.00
10 Gemini 3.8 FlashGoogle 1504 10,412±6.9 70.5 $4.125
11 Qwen3.8 Max (0902)Qwen 1503 7,434±7.6 63.5 $6.00
12 Claude Sonnet 5.5Anthropic 1502 1,914±13.9 88.9 $10.00
13 Muse Spark 1.3Meta 1500 4,884±9.0 79.7 $4.25
14 Kimi K3Moonshot AI 1496 8,333±7.2 61.8 $11.936
15 GLM 5.3Z.ai 1496 6,109±8.2 73.1 $1.40
16 MiMo-V2.5-ProXiaomi 1494 21,733±5.3 40.0 $0.609
17 Claude Sonnet 4.6Anthropic 1493 21,354±5.4 57.8 $15.00
18 MiMo-V2.6-FlashXiaomi 1493 2,917±11.3 61.9 $0.28
19 GLM 5.3 FlashZ.ai 1488 8,456±7.3 71.0 $0.50
20 DeepSeek V4.1 FlashDeepSeek 1488 3,257±10.6 77.2 $0.525
21 Muse Spark 1.2Meta 1487 2,442±12.2 65.1 $4.25
22 Gemini 3.7 FlashGoogle 1487 7,632±7.6 69.9 $3.75
23 Claude Opus 4.8Anthropic 1487 18,802±5.7 66.0 $25.00
24 Gemini 3.6 FlashGoogle 1484 10,120±6.8 48.3 $3.75
25 GPT-5.5OpenAI 1483 21,950±5.4 67.8 $30.00
26 Gemini 3.5 FlashGoogle 1483 13,780±6.0 42.8 $9.00
27 GLM 5.1Z.ai 1482 18,277±5.4 44.0 $4.40
28 Claude Opus 4.5Anthropic 1482 18,761±5.2 66.8 $25.00
29 GLM 5.2Z.ai 1481 13,844±6.0 57.7 $7.00
30 Gemini 3.1 Pro PreviewGoogle 1480 37,705±4.5 35.3 $12.00
31 Muse Spark 1.1Meta 1480 11,922±6.3 48.9 $4.25
32 GPT-5.4OpenAI 1479 19,714±5.6 37.7 $15.00
33 GPT-5.6 SolOpenAI 1477 10,914±6.6 76.7 $20.00
34 Claude Sonnet 4.5Anthropic 1471 21,152±4.8 59.1 $15.00
35 Kimi K2.6Moonshot AI 1471 12,470±6.5 44.2 $3.41
36 DeepSeek V4 Pro 0423DeepSeek 1466 17,839±5.8 35.3 $1.9105
37 GPT-5.6 TerraOpenAI 1465 11,785±6.4 69.0 $12.00
38 GPT-6.1 SolOpenAI 1465 1,655±15.0 88.7 $10.00
39 GPT-6 AstraOpenAI 1464 3,256±10.8 88.7 $50.00
40 Claude Sonnet 5Anthropic 1464 14,053±6.0 63.5 $10.00
41 Grok 4.5xAI 1464 10,720±6.7 75.5 $6.00
42 DeepSeek V4 Pro 0813DeepSeek 1463 2,490±12.0 54.6 $1.98
43 GPT-5.1OpenAI 1462 8,815±6.9 53.3 $10.00
44 Qwen3.8 27BQwen 1461 6,237±8.1 73.6 $1.35
45 MiMo-V2.5Xiaomi 1461 15,061±6.3 34.0 $0.238
46 Kimi K2.5Moonshot AI 1460 21,040±5.2 58.5 $2.50
47 Qwen3.7 PlusQwen 1460 12,134±6.4 Not benchmarked $1.28
48 GLM 5Z.ai 1460 8,278±7.0 59.7 $3.20
49 GPT-5.6 LunaOpenAI 1460 10,956±6.6 61.5 $1.20
50 Gemma 4 31BGoogle 1456 1,328±15.7 38.3 $0.40
51 GLM 5V TurboZ.ai 1456 2,950±11.0 28.8 $4.00
52 Qwen3.6 PlusQwen 1455 15,348±6.1 37.4 $1.95
53 Hy3Tencent 1455 3,247±10.5 43.9 $0.528
54 MiniMax M3MiniMax 1455 18,479±5.6 57.8 $1.20
55 Step 5 PreviewStepFun 1454 2,022±13.2 71.4 $2.70
56 Nemotron 3 UltraNVIDIA 1453 3,663±10.3 Not benchmarked $2.40
57 Gemini 2.5 ProGoogle 1453 28,958±4.1 37.0 $10.00
58 GLM 4.7Z.ai 1452 2,699±11.3 33.6 $1.98
59 Claude Opus 4.1Anthropic 1451 10,730±6.2 27.2 $75.00
60 R Beam 1.0 501B A23BReflection 1451 1,048±18.1 33.8 $1.00
61 Qwen3.5 397B A17BQwen 1450 28,085±4.7 28.6 $2.34
62 DeepSeek V4 Flash 0423DeepSeek 1449 16,217±6.0 54.6 $0.182
63 T InklingThinking Machines 1448 11,262±6.6 49.7 $4.05
64 Gemma 4 26B A4BGoogle 1448 1,276±15.8 22.8 $0.255
65 DeepSeek V3.2DeepSeek 1446 9,597±6.4 54.4 $0.42
66 Mistral Medium 3.5Mistral AI 1445 3,854±10.0 9.2 $7.50
67 GLM 4.6Z.ai 1445 8,251±6.9 52.2 $2.20
68 Grok 4.6xAI 1445 7,999±7.4 76.4 $6.00
69 DeepSeek V3.2 ExpDeepSeek 1444 2,060±12.9 55.0 $0.41
70 Mistral Large 4Mistral AI 1442 2,312±12.5 55.3 $4.18
71 MiniMax M2.7MiniMax 1441 24,068±5.0 28.4 $1.20
72 Gemini 3.5 Flash LiteGoogle 1440 9,644±7.0 33.6 $2.50
73 Qwen3 VL 235B A22B InstructQwen 1439 2,499±12.7 Not benchmarked $1.90
74 Claude Haiku 4.5Anthropic 1435 41,092±3.9 49.8 $5.00
75 Mistral Medium 3.1Mistral AI 1435 23,170±4.4 Not benchmarked $2.00
76 DeepSeek V3.1 TerminusDeepSeek 1432 729±21.4 Not benchmarked $1.00
77 Qwen3 235B A22B Thinking 2507Qwen 1432 1,717±14.1 Not benchmarked $3.50
78 Qwen3 Next 80B A3B InstructQwen 1431 5,217±8.3 Not benchmarked $1.10
79 Qwen3.5-122B-A10BQwen 1431 8,670±6.9 22.9 $2.08
80 GPT-5.2OpenAI 1431 13,037±5.9 58.9 $14.00
81 T Inkling SmallThinking Machines 1429 9,094±7.1 29.9 $1.20
82 GLM 4.5Z.ai 1429 4,831±8.7 64.2 $2.20
83 Qwen3.5-27BQwen 1429 8,340±7.0 22.6 $2.16
84 Grok 4.7xAI 1427 3,251±10.9 75.0 $6.00
85 Solar Pro 4Upstage 1427 5,924±8.3 24.4 $0.60
86 R1 0528DeepSeek 1422 3,106±10.8 71.4 $2.15
87 Step 3.5 FlashStepFun 1421 16,231±5.6 Not benchmarked $0.30
88 Gemini 2.5 FlashGoogle 1420 28,913±4.1 28.7 $2.50
89 GPT-6 SolOpenAI 1420 3,341±10.9 82.8 $10.00
90 GPT-5.4 MiniOpenAI 1420 19,855±5.6 28.2 $4.50
91 GPT-5OpenAI 1419 6,656±7.7 72.2 $10.00
92 Hy3 previewTencent 1416 2,316±12.5 22.4 $0.60
93 o3OpenAI 1411 12,104±6.1 70.7 $8.00
94 Qwen3.5-35B-A3BQwen 1411 8,844±6.8 7.7 $1.00
95 GPT-6 LunaOpenAI 1410 15,274±6.2 71.4 $0.50
96 Qwen3.5-FlashQwen 1409 18,206±5.6 6.2 $0.26
97 Qwen3 30B A3B Instruct 2507Qwen 1406 4,747±8.7 Not benchmarked $0.30
98 Gemini 3.1 Flash Lite PreviewGoogle 1403 19,972±5.5 8.1 $1.50
99 Grok 4.3xAI 1402 22,445±5.4 22.5 $2.50
100 Nemotron 3 SuperNVIDIA 1397 1,768±13.6 Not benchmarked $2.40
101 GPT-4.1OpenAI 1393 10,441±6.3 46.5 $8.00
102 GLM 4.6VZ.ai 1391 640±22.9 Not benchmarked $0.90
103 GLM 4.5 AirZ.ai 1388 6,579±7.4 Not benchmarked $0.85
104 GPT-5.4 NanoOpenAI 1387 19,932±5.7 Not benchmarked $1.25
105 Nova 2 LiteAmazon 1385 2,746±11.2 Not benchmarked $2.50
106 Qwen3 Next 80B A3B ThinkingQwen 1384 3,018±10.7 Not benchmarked $1.20
107 MiniMax M2.5MiniMax 1384 12,185±6.3 59.4 $0.95
108 GPT-5 MiniOpenAI 1383 5,601±8.2 59.8 $2.00
109 Qwen3 Coder 480B A35BQwen 1380 5,143±8.7 50.0 $1.00
110 MiniMax M2MiniMax 1376 1,649±14.3 45.3 $1.02
111 Nemotron 3 Nano 30B A3BNVIDIA 1376 3,356±10.3 Not benchmarked $0.24
112 o3 MiniOpenAI 1375 2,944±10.9 52.1 $4.40
113 Mistral Medium 3Mistral AI 1374 6,364±7.8 Not benchmarked $2.00
114 Mercury 2Inception 1374 722±21.3 0.0 $0.75
115 o1OpenAI 1372 4,238±9.4 61.7 $60.00
116 GLM 4.7 FlashZ.ai 1369 2,556±11.4 Not benchmarked $0.40
117 gpt-oss-120bOpenAI 1368 6,479±7.7 34.5 $0.60
118 o4 MiniOpenAI 1366 9,115±6.7 68.6 $4.40
119 GPT-4.1 MiniOpenAI 1362 7,670±7.2 28.5 $1.60
120 GLM 4.5VZ.ai 1357 1,008±18.0 Not benchmarked $1.80
121 Trinity Large ThinkingArcee AI 1356 9,948±7.0 5.4 $0.80
122 MiniMax M1MiniMax 1353 6,994±7.3 Not benchmarked $2.20
123 Gemma 3 27BGoogle 1353 8,933±6.6 4.9 $0.35
124 Qwen3 32BQwen 1350 729±20.6 40.0 $0.28
125 Granite 4.2 8BIBM Granite 1345 900±21.0 Not benchmarked $0.25
126 GPT-5 NanoOpenAI 1338 1,696±14.4 34.8 $0.40
127 Command ACohere 1335 11,260±6.0 Not benchmarked $10.00
128 Qwen-PlusQwen 1333 964±17.3 Not benchmarked $0.78
129 Qwen3 30B A3BQwen 1329 5,160±8.4 Not benchmarked $0.50
130 Gemma 3 12BGoogle 1318 700±20.6 Not benchmarked $0.15
131 Llama 4 MaverickMeta Llama 1295 7,690±7.3 18.1 $0.6525
132 Llama 4 ScoutMeta Llama 1293 5,734±8.2 9.1 $0.30
133 GPT-4o (2024-05-13)OpenAI 1293 19,579±6.4 Not benchmarked $15.00
134 GPT-4.1 NanoOpenAI 1293 1,062±17.0 8.9 $0.40
135 gpt-oss-20bOpenAI 1292 2,081±13.3 Not benchmarked $0.09
136 Mistral Small 3.1 24BMistral AI 1290 6,526±7.8 Not benchmarked $0.555
137 Qwen2.5 72B InstructQwen 1279 6,208±7.8 Not benchmarked $0.40
138 GPT-4o-mini (2024-07-18)OpenAI 1278 11,355±6.4 3.6 $0.60
139 GPT-4o (2024-08-06)OpenAI 1277 8,010±7.7 23.1 $10.00
140 Llama 3.3 70B InstructMeta Llama 1277 9,545±6.3 Not benchmarked $0.32
141 Qwen2.5 Coder 32B InstructQwen 1269 759±19.3 30.5 $1.00
142 Mistral LargeMistral AI 1268 4,452±8.6 Not benchmarked $6.00
143 GPT-4 TurboOpenAI 1266 17,922±7.0 Not benchmarked $30.00
144 Llama 3.1 70B InstructMeta Llama 1261 9,133±7.1 Not benchmarked $0.40
145 Nova Pro 1.0Amazon 1259 4,069±8.8 Not benchmarked $3.20
146 Gemma 3 4BGoogle 1253 764±20.3 Not benchmarked $0.05
147 Mistral Small 3Mistral AI 1243 2,355±11.7 Not benchmarked $0.08
148 Phi 4Microsoft 1230 3,804±9.4 Not benchmarked $0.14
149 Nova Lite 1.0Amazon 1226 3,136±10.2 Not benchmarked $0.24
150 GPT-4OpenAI 1209 9,838±8.8 22.4 $60.00
151 Gemma 2 27BGoogle 1206 12,606±6.3 Not benchmarked $0.65
152 Nova Micro 1.0Amazon 1203 3,050±10.2 Not benchmarked $0.14
153 Llama 3.1 8B InstructMeta Llama 1196 8,195±7.5 Not benchmarked $0.08
154 Command R (08-2024)Cohere 1194 1,735±13.2 Not benchmarked $0.60
155 Llama 3.2 3B InstructMeta Llama 1141 1,377±15.0 Not benchmarked $0.335
156 Llama 3.2 1B InstructMeta Llama 1087 1,437±14.9 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.