llm leaderboard

Best AI for design

Design here means producing an interface a person prefers, not writing about design.

Aldena runs these models inside your team's projects. See what each one costs.

148 of 148 ranked models
Ranked models
rankmodelvendorcompositepricein / outbenchmarkseloeloeloelo
1
Claude Opus 5.5 (max)anthropic/claude-opus-5.5:max
Anthropic81.9$4.00 / $20.004 of 4 benchmarks1818186917341751
2
Claude Sonnet 5.5 (xhigh)anthropic/claude-sonnet-5.5:xhigh
Anthropic81.6$2.00 / $10.004 of 4 benchmarks1773182217541744
3
GPT-6 Astra (max)openai/gpt-6-astra:max
OpenAI80.7$10.00 / $50.004 of 4 benchmarks1767178117841721
4
Claude Fable 5.1 (max)anthropic/claude-fable-5.1:max
Anthropic80.1$10.00 / $50.004 of 4 benchmarks1744178816851724
5
GPT-6.1 Sol (max)openai/gpt-6.1-sol:max
OpenAI80.1$2.00 / $10.004 of 4 benchmarks1743174717931697
6
Claude Opus 5 (max)anthropic/claude-opus-5:max
Anthropic78.3$5.00 / $25.004 of 4 benchmarks1691171316791664
7
Gemini 4 Carbon (high)google/gemini-4-carbon:high
Google78.1—3 of 4 benchmarks176817921771
8
GPT-6 Sol (max)openai/gpt-6-sol:max
OpenAI76.6$2.00 / $10.004 of 4 benchmarks1674167616781657
9
Claude Sonnet 5.5 (high)anthropic/claude-sonnet-5.5:high
Anthropic76.6$2.00 / $10.003 of 4 benchmarks171317481704
10
Qwen3.8 Max (0902)qwen/qwen3.8-max-0902
Qwen76.1$2.00 / $6.004 of 4 benchmarks1677169816551637
11
Kimi K3.1 Pro (max)moonshotai/kimi-k3.1-pro:max
MoonshotAI75.4$2.00 / $10.003 of 4 benchmarks168716991686
12
DeepSeek V4.1 Flash (max)deepseek/deepseek-v4.1-flash:max
DeepSeek75.4$0.051 / $0.5674 of 4 benchmarks1664167516961610
13
Muse Spark 1.3 (max)meta/muse-spark-1.3:max
Meta74.8$1.25 / $4.254 of 4 benchmarks1654166816531638
14
Gemini 4 Argon (high)google/gemini-4-argon:high
Google74.7$2.00 / $10.003 of 4 benchmarks168317031661
15
Qwen3.8 Maxqwen/qwen3.8-max
Qwen74.4—4 of 4 benchmarks1670167916261615
16
GPT-6.1 Terra (max)openai/gpt-6.1-terra:max
OpenAI74.3$1.00 / $5.003 of 4 benchmarks167816621744
17
Claude Opus 5 (high)anthropic/claude-opus-5:high
Anthropic72.6$5.00 / $25.003 of 4 benchmarks165616781650
18
Claude Fable 5 (high)anthropic/claude-fable-5:high
Anthropic72.1$10.00 / $50.004 of 4 benchmarks1625163216491621
19
Inkling 2 Smallthinkingmachines/inkling-2-small
Thinking Machines71.5—3 of 4 benchmarks164916611661
20
Kimi K3 (max)moonshotai/kimi-k3:max
MoonshotAI71.4$0.192 / $16.004 of 4 benchmarks1657167216411586
21
Muse Wonder (max)meta/muse-wonder:max
Meta71.3$4.00 / $20.003 of 4 benchmarks165016501666
22
Grok 4.7 (xhigh)x-ai/grok-4.7:xhigh
SpaceXAI70.6$2.00 / $6.004 of 4 benchmarks1640166216351591
23
MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash
Xiaomi70.5$0.14 / $0.284 of 4 benchmarks1636165316361594
24
Hy4 previewtencent/hy4-preview
Tencent69.6$0.7506 / $2.25093 of 4 benchmarks163216531624
25
MiMo-V2.6-Proxiaomi/mimo-v2.6-pro
Xiaomi69.5$0.435 / $0.874 of 4 benchmarks1624163716231595
26
Grok 4.6 (high)x-ai/grok-4.6:high
SpaceXAI69.4$2.00 / $6.004 of 4 benchmarks1618163116001607
27
Muse Spark 1.3 (xhigh)meta/muse-spark-1.3:xhigh
Meta69.0$1.25 / $4.254 of 4 benchmarks1621163016191599
28
GPT-5.6 Sol (xhigh)openai/gpt-5.6-sol:xhigh
OpenAI68.8$4.00 / $20.004 of 4 benchmarks1615161816121606
29
Qwen3.8 Flash Nextqwen/qwen3.8-flash-next
Qwen68.2—3 of 4 benchmarks162416541594
30
GLM 5.3 (max)z-ai/glm-5.3:max
Z.ai67.7$0.0699 / $1.403 of 4 benchmarks162216431595
31
Gemini 3.7 Flash (high)google/gemini-3.7-flash:high
Google66.4$0.75 / $3.754 of 4 benchmarks1588158415931600
32
GLM 5.3 Flashz-ai/glm-5.3-flash
Z.ai66.0$0.15 / $0.504 of 4 benchmarks1609162815761582
33
Gemini 3.8 Flash (high)google/gemini-3.8-flash:high
Google65.6$0.75 / $3.754 of 4 benchmarks1584157716091594
34
GLM 5.2 (max)z-ai/glm-5.2:max
Z.ai65.3$0.006 / $6.803 of 4 benchmarks160816211567
35
Claude Haiku 5.5 (high)anthropic/claude-haiku-5.5:high
Anthropic64.7$0.10 / $0.503 of 4 benchmarks159516151566
36
Qwen3.8 27Bqwen/qwen3.8-27b
Qwen64.4$0.15 / $1.8754 of 4 benchmarks1591159615711574
37
MiniMax M3.1 Flashminimax/minimax-m3.1-flash
MiniMax63.3$0.075 / $0.303 of 4 benchmarks158216021557
38
GPT-6 Luna (max)openai/gpt-6-luna:max
OpenAI63.1$0.10 / $0.504 of 4 benchmarks1577157515951567
39
DeepSeek V4 Pro 0423 (high)deepseek/deepseek-v4-pro:high
DeepSeek63.0$0.8837 / $1.76753 of 4 benchmarks158115861564
40
DeepSeek V4 Flash 0423 (high)deepseek/deepseek-v4-flash:high
DeepSeek62.7$0.091 / $0.1823 of 4 benchmarks158215851559
41
Mistral Large 4mistralai/mistral-large-4-0
Mistral61.3$0.70 / $2.103 of 4 benchmarks156315701554
42
Claude Opus 4.7 (high)anthropic/claude-opus-4.7:high
Anthropic61.1$5.00 / $25.004 of 4 benchmarks1557156015451579
43
Step 5 Preview (high)stepfun/step-5-preview:high
StepFun61.0$1.00 / $2.704 of 4 benchmarks1572158915451551
44
Grok 4.5x-ai/grok-4.5
SpaceXAI60.6$2.00 / $6.004 of 4 benchmarks1552155615651562
45
Claude Opus 4.7anthropic/claude-opus-4.7
Anthropic60.1$5.00 / $25.004 of 4 benchmarks1555155215511562
46
Claude Opus 4.8 (high)anthropic/claude-opus-4.8:high
Anthropic59.6$5.00 / $25.003 of 4 benchmarks155515561542
47
Claude Opus 4.6 (high)anthropic/claude-opus-4.6:high
Anthropic59.1$5.00 / $25.004 of 4 benchmarks1546153915591544
48
Gemini 3.6 Flash (high)google/gemini-3.6-flash:high
Google57.9$0.75 / $3.754 of 4 benchmarks1538152715471545
49
Muse Spark 1.1meta/muse-spark-1.1
Meta57.0$1.25 / $4.254 of 4 benchmarks1542153115391539
50
Claude Opus 4.8anthropic/claude-opus-4.8
Anthropic57.0$5.00 / $25.003 of 4 benchmarks153615371525
51
Muse Spark 1.2 (xhigh)meta/muse-spark-1.2:xhigh
Meta57.0$1.25 / $4.253 of 4 benchmarks153415251542
52
Claude Sonnet 5 (high)anthropic/claude-sonnet-5:high
Anthropic56.7$2.00 / $10.004 of 4 benchmarks1539153915261539
53
Claude Opus 4.6anthropic/claude-opus-4.6
Anthropic56.5$5.00 / $25.004 of 4 benchmarks1537152815471538
54
Qwen3.7 Maxqwen/qwen3.7-max
Qwen54.4$1.475 / $4.4253 of 4 benchmarks151515121514
55
Muse Spark Open (max)meta/muse-spark-open:max
Meta54.1$1.25 / $4.253 of 4 benchmarks152815261476
56
GPT-5.6 Terra (xhigh)openai/gpt-5.6-terra:xhigh
OpenAI54.0$2.00 / $12.004 of 4 benchmarks1520151115231539
57
Claude Sonnet 4.6anthropic/claude-sonnet-4.6
Anthropic53.8$3.00 / $15.004 of 4 benchmarks1522151115261529
58
GPT-5.5 (xhigh)openai/gpt-5.5:xhigh
OpenAI53.5$5.00 / $30.004 of 4 benchmarks1512150515431531
59
GLM 5.1z-ai/glm-5.1
Z.ai52.7$1.40 / $4.403 of 4 benchmarks150815041510
60
Hy3tencent/hy3
Tencent52.3$0.0825 / $0.333 of 4 benchmarks150615071490
61
GPT-5.6 Luna (xhigh)openai/gpt-5.6-luna:xhigh
OpenAI51.9$0.20 / $1.204 of 4 benchmarks1519150815221517
62
Claude Opus 4.5 (high 32K)anthropic/claude-opus-4.5:high-32k
Anthropic51.5$5.00 / $25.003 of 4 benchmarks149314761510
63
Seed 2.1 Pro Previewbytedance/seed-2.1-pro-preview
ByteDance50.9—4 of 4 benchmarks1519151514901505
64
GPT-5.5 (high)openai/gpt-5.5:high
OpenAI50.7$5.00 / $30.004 of 4 benchmarks1487147715341519
65
Qwen3.6 Max Previewqwen/qwen3.6-max-preview
Qwen50.4—3 of 4 benchmarks148214731492
66
Kimi K2.6moonshotai/kimi-k2.6
MoonshotAI50.3$0.65 / $3.414 of 4 benchmarks1509150414861522
67
Gemini 3.5 Flash (medium)google/gemini-3.5-flash:medium
Google49.6$1.50 / $9.004 of 4 benchmarks1490148015221506
68
Gemini 3.5 Flash (high)google/gemini-3.5-flash:high
Google48.9$1.50 / $9.004 of 4 benchmarks1500149515131500
69
Claude Opus 4.5anthropic/claude-opus-4.5
Anthropic48.8$5.00 / $25.003 of 4 benchmarks146914561482
70
MiMo-V2.5-Proxiaomi/mimo-v2.5-pro
Xiaomi48.6$0.48 / $1.803 of 4 benchmarks147914681471
71
GPT-5.4 (high)openai/gpt-5.4:high
OpenAI48.0$2.50 / $15.003 of 4 benchmarks146214511474
72
Kimi K2.7 Codemoonshotai/kimi-k2.7-code
MoonshotAI46.6$0.60 / $3.504 of 4 benchmarks1474147414541512
73
GPT-5.4 (medium)openai/gpt-5.4:medium
OpenAI46.3$2.50 / $15.003 of 4 benchmarks144214271476
74
GPT-5.5openai/gpt-5.5
OpenAI46.1$5.00 / $30.004 of 4 benchmarks1455144414911504
75
MiniMax M3minimax/minimax-m3
MiniMax45.9$0.30 / $1.204 of 4 benchmarks1484147914601474
76
DeepSeek V4 Pro 0423deepseek/deepseek-v4-pro
DeepSeek45.7$0.8837 / $1.76753 of 4 benchmarks144814411434
77
Beam 1.0 501B A23Breflection/beam-1.0-501b-a23b
Reflection44.9$0.30 / $1.003 of 4 benchmarks143814431431
78
Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview
Google44.6$2.00 / $12.004 of 4 benchmarks1446143814891480
79
GPT-5.2openai/gpt-5.2
OpenAI44.4$1.75 / $14.002 of 4 benchmarks14201429
80
Qwen3.6 Plusqwen/qwen3.6-plus
Qwen44.1$0.325 / $1.954 of 4 benchmarks1462145314631473
81
GLM 5z-ai/glm-5
Z.ai44.1$1.00 / $3.203 of 4 benchmarks143514241449
82
GPT-5 (medium)openai/gpt-5:medium
OpenAI43.7$1.25 / $10.002 of 4 benchmarks14171425
83
Mimo v2 Proxiaomi/mimo-v2-pro
Xiaomi43.6—3 of 4 benchmarks143414251440
84
MiMo-V2.5xiaomi/mimo-v2.5
Xiaomi43.5$0.40 / $2.003 of 4 benchmarks143814231429
85
GLM 4.7z-ai/glm-4.7
Z.ai43.5$0.60 / $2.203 of 4 benchmarks143514181446
86
Gemini 3 Flash Previewgoogle/gemini-3-flash-preview
Google41.7$0.50 / $3.004 of 4 benchmarks1440142814441451
87
Gemini 3 Progoogle/gemini-3-pro
Google40.4—4 of 4 benchmarks1441139614721441
88
GPT-5.1 (medium)openai/gpt-5.1:medium
OpenAI40.4$1.25 / $10.002 of 4 benchmarks13971406
89
Inklingthinkingmachines/inkling
Thinking Machines40.3$1.00 / $4.053 of 4 benchmarks141414041405
90
Kimi K2.5 (thinking)moonshotai/kimi-k2.5:thinking
MoonshotAI39.5$0.532 / $3.3254 of 4 benchmarks1436142814271439
91
Gemini 3.5 Flash Litegoogle/gemini-3.5-flash-lite
Google39.3$0.30 / $2.503 of 4 benchmarks143514081449
92
Claude Opus 4.1anthropic/claude-opus-4.1
Anthropic39.2$15.00 / $75.002 of 4 benchmarks13891397
93
Qwen3.5 397B A17Bqwen/qwen3.5-397b-a17b
Qwen39.0$0.60 / $3.603 of 4 benchmarks140013871409
94
GPT-5.4 Mini (high)openai/gpt-5.4-mini:high
OpenAI38.4$0.75 / $4.503 of 4 benchmarks139813861420
95
MiniMax M2.7minimax/minimax-m2.7
MiniMax37.9$0.21 / $0.843 of 4 benchmarks139813871401
96
GPT-5.3-Codexopenai/gpt-5.3-codex
OpenAI37.5$1.75 / $14.004 of 4 benchmarks1406139414211443
97
MiniMax M2.5minimax/minimax-m2.5
MiniMax36.9$0.30 / $1.203 of 4 benchmarks138613731414
98
Claude Sonnet 4.5 (high 32K)anthropic/claude-sonnet-4.5:high-32k
Anthropic36.9$3.00 / $15.003 of 4 benchmarks139413861398
99
GPT-5.4openai/gpt-5.4
OpenAI36.6$2.50 / $15.004 of 4 benchmarks1400138714241441
100
GPT-5.1 (high)openai/gpt-5.1:high
OpenAI35.9$1.25 / $10.001 of 4 benchmarks1419
101
Claude Sonnet 4.5anthropic/claude-sonnet-4.5
Anthropic35.6$3.00 / $15.003 of 4 benchmarks138513821389
102
MiniMax M2.1minimax/minimax-m2.1
MiniMax35.5—3 of 4 benchmarks138413651398
103
Kimi K2.5 Instantmoonshotai/kimi-k2.5-instant
Moonshot AI34.9—4 of 4 benchmarks1406139214111411
104
Inkling Smallthinkingmachines/inkling-small
Thinking Machines34.3$0.45 / $1.204 of 4 benchmarks1407139613951410
105
Nemotron 3.5 Super VLnvidia/nemotron-3.5-super-vl
NVIDIA34.2$0.15 / $0.474 of 4 benchmarks1399139114201391
106
Grok 4.20 Beta 0309 (reasoning)x-ai/grok-4.20-beta-0309:reasoning
SpaceXAI34.2—3 of 4 benchmarks137613681370
107
Gemma 4 31Bgoogle/gemma-4-31b-it
Google34.0$0.20 / $0.403 of 4 benchmarks136913611379
108
Solar Pro 4upstage/solar-pro4
Upstage33.6$0.15 / $0.603 of 4 benchmarks136913611371
109
GLM 4.6z-ai/glm-4.6
Z.ai33.4$0.50 / $2.002 of 4 benchmarks13421351
110
DeepSeek V3.2 (thinking)deepseek/deepseek-v3.2:thinking
DeepSeek32.7$0.26 / $0.383 of 4 benchmarks136313501370
111
GPT-5.1-Codexopenai/gpt-5.1-codex
OpenAI32.7$1.25 / $10.002 of 4 benchmarks13381346
112
Gemini 3 Flash Preview (minimal)google/gemini-3-flash-preview:minimal
Google32.2$0.50 / $3.004 of 4 benchmarks1384137413951431
113
Qwen3.5-27Bqwen/qwen3.5-27b
Qwen32.0$0.26 / $2.603 of 4 benchmarks135713451384
114
GLM 5V Turboz-ai/glm-5v-turbo
Z.ai31.9$1.20 / $4.004 of 4 benchmarks1399138213581427
115
Qwen3.5-122B-A10Bqwen/qwen3.5-122b-a10b
Qwen31.6$0.26 / $2.083 of 4 benchmarks136013511352
116
Gemma 4 26B A4B google/gemma-4-26b-a4b-it
Google31.6$0.0765 / $0.2553 of 4 benchmarks135913501362
117
Hy3 previewtencent/hy3-preview
Tencent31.3$0.18 / $0.603 of 4 benchmarks135913501355
118
Muse Glimmer 30Bmeta/muse-glimmer-30b
Meta31.0$0.30 / $1.203 of 4 benchmarks135513471364
119
MiniMax M2minimax/minimax-m2
MiniMax30.6—2 of 4 benchmarks12991308
120
Laguna M.1poolside/laguna-m.1
Poolside29.2—3 of 4 benchmarks134913381333
121
DeepSeek V3.2 Expdeepseek/deepseek-v3.2-exp
DeepSeek29.1$0.27 / $0.412 of 4 benchmarks12701279
122
GPT-5.2-Codexopenai/gpt-5.2-codex
OpenAI28.9$1.75 / $14.003 of 4 benchmarks133813291348
123
Mimo v2 Flashxiaomi/mimo-v2-flash
Xiaomi28.5—3 of 4 benchmarks133113121353
124
KAT-Coder-Pro V1kwaipilot/kat-coder-pro-v1
Kwaipilot28.4—2 of 4 benchmarks12571266
125
Grok 4.3x-ai/grok-4.3
SpaceXAI27.9$1.25 / $2.504 of 4 benchmarks1357135213671375
126
Claude Haiku 4.5anthropic/claude-haiku-4.5
Anthropic27.6$1.00 / $5.003 of 4 benchmarks133013251318
127
GPT-5.1-Codex-Miniopenai/gpt-5.1-codex-mini
OpenAI27.5$0.25 / $2.002 of 4 benchmarks12451253
128
DeepSeek V3.2deepseek/deepseek-v3.2
DeepSeek27.4$0.26 / $0.383 of 4 benchmarks132613331306
129
Kimi K2 Thinking Turbomoonshotai/kimi-k2-thinking-turbo
Moonshot AI27.3—3 of 4 benchmarks132213131325
130
Grok 4.1 (thinking)x-ai/grok-4.1:thinking
SpaceXAI25.8—2 of 4 benchmarks12171226
131
Mimo v2 Flash (thinking)xiaomi/mimo-v2-flash:thinking
Xiaomi25.7—3 of 4 benchmarks129212281344
132
Laguna XS.2poolside/laguna-xs.2
Poolside25.6—3 of 4 benchmarks130312951285
133
Mistral Medium 3.5mistralai/mistral-medium-3-5
Mistral25.5$1.50 / $7.503 of 4 benchmarks126212521308
134
GPT-5.1openai/gpt-5.1
OpenAI25.2$1.25 / $10.004 of 4 benchmarks1342130713641349
135
Qwen3 Coder 480B A35b Instructqwen/qwen3-coder-480b-a35b-instruct
Qwen25.2—3 of 4 benchmarks127512581286
136
Grok Code Fast 1x-ai/grok-code-fast-1
SpaceXAI24.4—2 of 4 benchmarks11641171
137
Qwen3.5-35B-A3Bqwen/qwen3.5-35b-a3b
Qwen24.3$0.14 / $1.003 of 4 benchmarks125312361304
138
Grok 4 Fast (reasoning)x-ai/grok-4-fast:reasoning
SpaceXAI24.1—2 of 4 benchmarks11621170
139
Devstral Medium 2507mistralai/devstral-medium-2507
Mistral AI23.7—2 of 4 benchmarks10721078
140
Qwen3.5-Flashqwen/qwen3.5-flash-02-23
Qwen23.6$0.065 / $0.263 of 4 benchmarks124412291290
141
Grok 4.1 Fast (reasoning)x-ai/grok-4-1-fast:reasoning
SpaceXAI22.3—3 of 4 benchmarks124212211254
142
Mistral Large 3mistralai/mistral-large-3
Mistral AI22.3—3 of 4 benchmarks123312411362
143
Trinity Large Thinkingarcee-ai/trinity-large-thinking
Arcee AI22.1—3 of 4 benchmarks123612301232
144
Gemini 2.5 Progoogle/gemini-2.5-pro
Google21.1$1.25 / $10.003 of 4 benchmarks122812361285
145
Granite 4.1 8Bibm-granite/granite-4.1-8b
IBM20.4—3 of 4 benchmarks119311801208
146
Devstral 2 2512mistralai/devstral-2512
Mistral20.4$0.40 / $2.003 of 4 benchmarks119611471217
147
Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview
Google20.2$0.25 / $1.504 of 4 benchmarks1256124412751338
148
Mercury 2inception/mercury-2
Inception20.0—3 of 4 benchmarks117111601199
How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 2 of the 4 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

Data sources

Turn a source off to drop every benchmark it feeds and re-rank the board in your browser. Your choice follows you across the leaderboard pages.

  • LMArenaCC BY 4.0

    Arena ratings by LMArena, from the public leaderboard dataset.