LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

LMArena Text: instruction following leaderboard

Claude Opus 5.5 leads
1537
models measured
156
lead over #2
+3 Elo
median
1405

Leader on LMArena Text: instruction following · read October 21

  1. 1537 Claude Opus 5.5
  2. 1534 Gemini 4 Argon
  3. 1521 Claude Opus 4.6

On October 21, 2026, 156 of our catalog's models had a LMArena Text: instruction following result, and Anthropic Claude Opus 5.5 led with 1537.

Prompts with explicit instructions the answer has to follow. 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: instruction following results for models in the catalog
# Model Result VotesInterval (±) Evidence Coding index Output $/Mtok
1 Claude Opus 5.5Anthropic 1537 3,298±11.1 92.2 $20.00
2 Gemini 4 ArgonGoogle 1534 1,748±14.6 80.5 $10.00
3 Claude Opus 4.6Anthropic 1521 26,659±5.0 66.9 $25.00
4 Claude Fable 5.1Anthropic 1518 4,693±9.4 89.1 $50.00
5 Claude Opus 5Anthropic 1518 11,699±6.7 81.8 $25.00
6 Claude Fable 5Anthropic 1502 15,293±5.9 82.1 $50.00
7 Claude Opus 4.7Anthropic 1498 22,751±5.4 51.1 $25.00
8 Claude Sonnet 5.5Anthropic 1492 2,756±11.7 88.9 $10.00
9 Gemini 3.8 FlashGoogle 1492 14,854±6.3 70.5 $4.125
10 MiMo-V2.6-ProXiaomi 1491 1,391±15.8 72.6 $0.87
11 Muse WonderMeta 1491 1,797±14.2 64.5 $20.00
12 Kimi K3Moonshot AI 1485 11,432±6.4 61.8 $11.936
13 Gemini 3.7 FlashGoogle 1484 10,486±6.9 69.9 $3.75
14 GPT-5.6 SolOpenAI 1483 14,557±6.1 76.7 $20.00
15 Qwen3.8 Max (0902)Qwen 1481 10,450±6.7 63.5 $6.00
16 GLM 5.3Z.ai 1479 8,022±7.4 73.1 $1.40
17 Muse Spark 1.3Meta 1479 6,905±7.8 79.7 $4.25
18 GPT-5.5OpenAI 1478 24,002±5.3 67.8 $30.00
19 Claude Opus 4.5Anthropic 1478 9,886±6.6 66.8 $25.00
20 GLM 5.3 FlashZ.ai 1478 12,492±6.5 71.0 $0.50
21 Claude Opus 4.8Anthropic 1476 22,656±5.3 66.0 $25.00
22 Claude Sonnet 4.6Anthropic 1476 23,706±5.2 57.8 $15.00
23 MiMo-V2.5-ProXiaomi 1475 25,875±5.0 40.0 $0.609
24 DeepSeek V4.1 FlashDeepSeek 1473 4,725±9.1 77.2 $0.525
25 GPT-5.4OpenAI 1469 21,863±5.4 37.7 $15.00
26 Gemini 3.5 FlashGoogle 1467 17,393±5.6 42.8 $9.00
27 Gemini 3.6 FlashGoogle 1467 13,098±6.3 48.3 $3.75
28 GPT-6.1 SolOpenAI 1466 2,278±12.9 88.7 $10.00
29 GLM 5.2Z.ai 1466 17,693±5.6 57.7 $7.00
30 Gemini 3.1 Pro PreviewGoogle 1466 44,333±4.2 35.3 $12.00
31 Muse Spark 1.2Meta 1462 3,379±10.6 65.1 $4.25
32 Claude Sonnet 4.5Anthropic 1459 24,365±4.6 59.1 $15.00
33 Muse Spark 1.1Meta 1457 16,036±5.8 48.9 $4.25
34 GPT-5.6 TerraOpenAI 1456 16,102±5.9 69.0 $12.00
35 GPT-6 AstraOpenAI 1455 4,190±9.6 88.7 $50.00
36 GLM 5.1Z.ai 1453 22,539±5.0 44.0 $4.40
37 Claude Sonnet 5Anthropic 1452 17,926±5.6 63.5 $10.00
38 MiMo-V2.6-FlashXiaomi 1452 4,184±9.6 61.9 $0.28
39 Kimi K2.6Moonshot AI 1451 13,198±6.3 44.2 $3.41
40 DeepSeek V4 Pro 0813DeepSeek 1448 3,482±10.3 54.6 $1.98
41 DeepSeek V4 Pro 0423DeepSeek 1445 20,026±5.6 35.3 $1.9105
42 Grok 4.5xAI 1445 14,519±6.1 75.5 $6.00
43 GPT-5.1OpenAI 1443 11,162±6.3 53.3 $10.00
44 Qwen3.7 PlusQwen 1439 15,018±5.9 Not benchmarked $1.28
45 Step 5 PreviewStepFun 1438 3,099±10.9 71.4 $2.70
46 GPT-5.6 LunaOpenAI 1437 14,603±6.1 61.5 $1.20
47 Gemini 2.5 ProGoogle 1437 35,196±3.8 37.0 $10.00
48 Qwen3.8 27BQwen 1436 8,906±7.1 73.6 $1.35
49 Claude Opus 4.1Anthropic 1435 13,212±5.7 27.2 $75.00
50 MiMo-V2.5Xiaomi 1435 16,053±6.1 34.0 $0.238
51 Gemma 4 31BGoogle 1433 1,798±13.6 38.3 $0.40
52 MiniMax M3MiniMax 1433 22,903±5.2 57.8 $1.20
53 Grok 4.6xAI 1431 10,995±6.7 76.4 $6.00
54 Kimi K2.5Moonshot AI 1431 24,195±4.8 58.5 $2.50
55 R Beam 1.0 501B A23BReflection 1430 1,557±15.0 33.8 $1.00
56 GLM 5Z.ai 1428 9,597±6.5 59.7 $3.20
57 Hy3Tencent 1426 4,617±9.1 43.9 $0.528
58 T InklingThinking Machines 1425 15,291±6.0 49.7 $4.05
59 Qwen3.6 PlusQwen 1424 16,550±5.9 37.4 $1.95
60 Qwen3.5 397B A17BQwen 1424 32,566±4.4 28.6 $2.34
61 GLM 5V TurboZ.ai 1423 3,701±10.0 28.8 $4.00
62 Mistral Large 4Mistral AI 1422 3,262±10.7 55.3 $4.18
63 DeepSeek V4 Flash 0423DeepSeek 1421 17,950±5.8 54.6 $0.182
64 Gemma 4 26B A4BGoogle 1420 1,736±13.5 22.8 $0.255
65 Gemini 3.5 Flash LiteGoogle 1419 12,907±6.3 33.6 $2.50
66 Nemotron 3 UltraNVIDIA 1418 3,902±9.9 Not benchmarked $2.40
67 Mistral Medium 3.5Mistral AI 1415 3,832±9.9 9.2 $7.50
68 Claude Haiku 4.5Anthropic 1414 48,438±3.7 49.8 $5.00
69 GPT-6 SolOpenAI 1413 4,626±9.3 82.8 $10.00
70 DeepSeek V3.2DeepSeek 1413 13,557±5.7 54.4 $0.42
71 GLM 4.7Z.ai 1412 3,304±10.3 33.6 $1.98
72 GLM 4.6Z.ai 1410 10,285±6.3 52.2 $2.20
73 GPT-5.2OpenAI 1409 15,296±5.5 58.9 $14.00
74 GPT-6 LunaOpenAI 1408 21,675±5.8 71.4 $0.50
75 Grok 4.7xAI 1406 4,502±9.5 75.0 $6.00
76 Qwen3 VL 235B A22B InstructQwen 1406 2,946±11.3 Not benchmarked $1.90
77 GPT-5.4 MiniOpenAI 1405 21,732±5.4 28.2 $4.50
78 Gemini 2.5 FlashGoogle 1405 34,833±3.8 28.7 $2.50
79 MiniMax M2.7MiniMax 1405 29,021±4.7 28.4 $1.20
80 GLM 4.5Z.ai 1404 6,166±7.9 64.2 $2.20
81 DeepSeek V3.1 TerminusDeepSeek 1404 858±20.1 Not benchmarked $1.00
82 DeepSeek V3.2 ExpDeepSeek 1401 3,389±10.2 55.0 $0.41
83 Qwen3.5-122B-A10BQwen 1399 9,572±6.6 22.9 $2.08
84 T Inkling SmallThinking Machines 1399 12,430±6.4 29.9 $1.20
85 Mistral Medium 3.1Mistral AI 1398 27,630±4.1 Not benchmarked $2.00
86 Qwen3.5-27BQwen 1393 9,231±6.6 22.6 $2.16
87 Qwen3 Next 80B A3B InstructQwen 1388 6,416±7.5 Not benchmarked $1.10
88 GPT-5OpenAI 1388 8,372±7.1 72.2 $10.00
89 Qwen3 235B A22B Thinking 2507Qwen 1386 2,094±12.4 Not benchmarked $3.50
90 Step 3.5 FlashStepFun 1385 19,314±5.2 Not benchmarked $0.30
91 Hy3 previewTencent 1383 2,336±12.5 22.4 $0.60
92 R1 0528DeepSeek 1382 4,015±9.5 71.4 $2.15
93 Qwen3.5-35B-A3BQwen 1379 9,945±6.4 7.7 $1.00
94 Gemini 3.1 Flash Lite PreviewGoogle 1377 21,488±5.3 8.1 $1.50
95 Solar Pro 4Upstage 1375 8,359±7.3 24.4 $0.60
96 Qwen3.5-FlashQwen 1374 20,285±5.3 6.2 $0.26
97 o3OpenAI 1368 15,500±5.5 70.7 $8.00
98 o1OpenAI 1367 10,246±6.5 61.7 $60.00
99 GPT-4.1OpenAI 1367 13,294±5.7 46.5 $8.00
100 Grok 4.3xAI 1366 25,125±5.2 22.5 $2.50
101 Qwen3 30B A3B Instruct 2507Qwen 1363 5,970±7.8 Not benchmarked $0.30
102 GPT-5.4 NanoOpenAI 1362 21,465±5.5 Not benchmarked $1.25
103 GPT-5 MiniOpenAI 1357 7,013±7.5 59.8 $2.00
104 Qwen3 Coder 480B A35BQwen 1355 6,201±7.8 50.0 $1.00
105 GLM 4.5 AirZ.ai 1354 8,204±6.8 Not benchmarked $0.85
106 MiniMax M2.5MiniMax 1353 13,767±5.9 59.4 $0.95
107 GLM 4.6VZ.ai 1350 753±21.1 Not benchmarked $0.90
108 Nemotron 3 SuperNVIDIA 1347 2,154±12.4 Not benchmarked $2.40
109 Qwen3 Next 80B A3B ThinkingQwen 1344 3,536±9.8 Not benchmarked $1.20
110 Mistral Medium 3Mistral AI 1338 7,949±7.1 Not benchmarked $2.00
111 o3 MiniOpenAI 1337 6,681±7.7 52.1 $4.40
112 Nova 2 LiteAmazon 1335 3,419±10.1 Not benchmarked $2.50
113 GPT-4.1 MiniOpenAI 1333 10,004±6.5 28.5 $1.60
114 MiniMax M2MiniMax 1329 2,023±12.8 45.3 $1.02
115 Mercury 2Inception 1328 886±19.1 0.0 $0.75
116 GLM 4.7 FlashZ.ai 1327 3,304±10.1 Not benchmarked $0.40
117 Trinity Large ThinkingArcee AI 1326 10,340±6.8 5.4 $0.80
118 Gemma 3 27BGoogle 1321 12,501±5.7 4.9 $0.35
119 o4 MiniOpenAI 1320 11,883±6.0 68.6 $4.40
120 gpt-oss-120bOpenAI 1318 7,865±7.0 34.5 $0.60
121 MiniMax M1MiniMax 1312 8,643±6.7 Not benchmarked $2.20
122 GLM 4.5VZ.ai 1311 1,294±15.8 Not benchmarked $1.80
123 Command ACohere 1309 15,575±5.3 Not benchmarked $10.00
124 GPT-5 NanoOpenAI 1306 2,025±13.1 34.8 $0.40
125 Qwen3 32BQwen 1305 858±18.7 40.0 $0.28
126 Nemotron 3 Nano 30B A3BNVIDIA 1303 4,370±9.1 Not benchmarked $0.24
127 Qwen-PlusQwen 1303 2,249±12.1 Not benchmarked $0.78
128 Granite 4.2 8BIBM Granite 1300 1,137±18.6 Not benchmarked $0.25
129 Gemma 3 12BGoogle 1299 1,145±16.1 Not benchmarked $0.15
130 Qwen3 30B A3BQwen 1283 6,135±7.8 Not benchmarked $0.50
131 GPT-4o (2024-05-13)OpenAI 1278 43,766±5.0 Not benchmarked $15.00
132 Llama 4 MaverickMeta Llama 1267 10,499±6.5 18.1 $0.6525
133 GPT-4o (2024-08-06)OpenAI 1267 18,305±5.8 23.1 $10.00
134 GPT-4.1 NanoOpenAI 1266 2,015±12.4 8.9 $0.40
135 Mistral Small 3.1 24BMistral AI 1263 8,046±7.1 Not benchmarked $0.555
136 GPT-4o-mini (2024-07-18)OpenAI 1258 26,707±4.8 3.6 $0.60
137 Qwen2.5 72B InstructQwen 1254 16,363±5.6 Not benchmarked $0.40
138 GPT-4 TurboOpenAI 1249 36,297±5.5 Not benchmarked $30.00
139 Mistral LargeMistral AI 1249 10,971±6.0 Not benchmarked $6.00
140 Llama 4 ScoutMeta Llama 1248 7,475±7.3 9.1 $0.30
141 Llama 3.3 70B InstructMeta Llama 1242 18,781±5.0 Not benchmarked $0.32
142 Gemma 3 4BGoogle 1239 1,233±16.1 Not benchmarked $0.05
143 gpt-oss-20bOpenAI 1236 2,541±12.2 Not benchmarked $0.09
144 Nova Pro 1.0Amazon 1235 9,525±6.3 Not benchmarked $3.20
145 Llama 3.1 70B InstructMeta Llama 1231 21,910±5.3 Not benchmarked $0.40
146 Qwen2.5 Coder 32B InstructQwen 1223 2,227±11.8 30.5 $1.00
147 Mistral Small 3Mistral AI 1214 5,485±8.1 Not benchmarked $0.08
148 Gemma 2 27BGoogle 1206 29,545±4.8 Not benchmarked $0.65
149 Nova Lite 1.0Amazon 1205 7,809±7.1 Not benchmarked $0.24
150 Phi 4Microsoft 1201 9,162±6.6 Not benchmarked $0.14
151 GPT-4OpenAI 1201 18,087±7.2 22.4 $60.00
152 Command R (08-2024)Cohere 1197 4,024±9.3 Not benchmarked $0.60
153 Nova Micro 1.0Amazon 1174 7,716±7.1 Not benchmarked $0.14
154 Llama 3.1 8B InstructMeta Llama 1159 19,781±5.5 Not benchmarked $0.08
155 Llama 3.2 3B InstructMeta Llama 1089 3,171±10.7 Not benchmarked $0.335
156 Llama 3.2 1B InstructMeta Llama 1031 3,248±11.1 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.