Leaderboards
Models
Blog
Methodology
DesignArena by Intelligence
Code
Office
Image
Video
Audio
Code
Office
Image
Video
Audio
DesignArena by Intelligence

Overall Preference vs Speed

Generation Time (slower →)Elo Rating (better →)

Overall Preference vs Price

Price per 1M Tokens (Log scale)Elo Rating (better →)

DesignArena

The largest crowdsourced benchmark for AI-generated design.

© 2026 DesignArena by Intelligence

Leaderboards

  • All Leaderboards
  • Overall Agentic
  • Overall HTML

Build with AI

  • Start Building
  • Website
  • Web Apps
  • Fullstack
  • Mobile App
  • Game
  • 3D Asset
  • Image
  • Video
  • SVG
  • Slides
  • Text-to-Speech

Code Leaderboards

  • Website (HTML, Non-Agentic)
  • Website (Frontend, Agentic)
  • Website (Fullstack, Agentic)
  • Builder
  • Image to HTML
  • Video to Website
  • Image to Frontend
  • Agentic Image to HTML
  • Game Dev (HTML, Non-Agentic)
  • Game Dev (Three.js, Agentic)
  • Game Dev (Godot, Agentic)
  • React Native
  • Android (Kotlin)
  • 3D Design (HTML, Non-Agentic)
  • Blender
  • SVG
  • ASCII Art
  • Code Art
  • Code to Video

Office Leaderboards

  • Python-PPTX Slides
  • HTML Slides
  • Proprietary Harness Slides
  • Spreadsheet
  • PDF
  • LifeArena

Image Leaderboards

  • Image
  • Image Editing
  • Single Image Editing
  • Reference to Image

Video Leaderboards

  • Video
  • Image to Video
  • Multi-Image to Video
  • Multimodal to Video
  • Video Editing

Audio Leaderboards

  • Text-to-Speech
  • Audio Realism Bench
  • Speech-to-Speech

Company

  • Methodology
  • Blog
  • Careers
  • Changelog
  • Contact

Resources

  • All Models
  • System Prompts
  • Security
  • Privacy Policy
  • Terms of Use
Code
Office
Image
Video
Audio
Code
Office
Image
Video
Audio
DesignArena by Intelligence

Overall Preference vs Speed

Generation Time (slower →)Elo Rating (better →)

Overall Preference vs Price

Price per 1M Tokens (Log scale)Elo Rating (better →)

Overall

One ranking across the agentic and HTML boards.

How the harness works→
Overall Agentic
DesignArena by Intelligence
Claude Opus 5.5 (high)
1447

Claude Opus 5.5

(high)

GPT-6 Astra (max)
1399

GPT-6 Astra

(max)

Claude Fable 5.2
1371

Claude Fable 5.2

GPT-6.1 Sol (max)
1369

GPT-6.1 Sol

(max)

Claude Fable 5.1
1332

Claude Fable 5.1

Inkling 2 Small
1309

Inkling 2 Small

Gemini 4 Carbon
1299

Gemini 4 Carbon

Kimi K3.1 Pro
1294

Kimi K3.1 Pro

Claude Sonnet 5.5
1287

Claude Sonnet 5.5

Muse Spark 1.3 Max
1277

Muse Spark 1.3 Max

DeepSeek-V4.1-Flash (1017)
1276

DeepSeek-V4.1-Flash

(1017)

Gemini 4 Argon
1271

Gemini 4 Argon

#76
Mistral Medium 3.5
888

Mistral Medium 3.5

Evaluates model ability across a composite of agentic tasks: Frontend Web App, Full Stack, Agentic Game Dev, Godot, React Native, Android, Slides, and Spreadsheet. Each output is built by a coding agent across multiple files and tool calls. Results aggregate real-world user preferences across these categories to provide an overall view of agentic performance.

Overall Agentic Preference vs Speed

Fast Generation, High Elo
Generation Time (slower →)Elo Rating (better →)1.7m16.7m96411941491Claude Opus 5.5 (high)Muse Spark 1.3 MaxDeepSeek-V4.1-Flash (1017)Claude Fable 5Muse Spark 1.3 (xhigh)Claude Opus 4.8Gemini 3.5 FlashGPT-5.5Gemini 3.1 Pro PreviewGemini 3 Pro PreviewGrok 4.3

Overall Agentic Preference vs Price

Low Cost, High Elo
Average Cost per Generation, measured (Log scale)Elo Rating (better →)$0.03$0.05$0.10$0.25$0.50$1$2.5$5$10$2599512361488Claude Opus 5.5 (high)GPT-6.1 Sol (max)Inkling 2 SmallGPT-6.1 SolMiniMax M3.1 FlashLing 3.2 FlashGPT-6 Luna
Models
Claude Opus 5.5 (high)
Claude Opus 5.5 (high)
GPT-6.1 Sol (max)
GPT-6.1 Sol (max)
Inkling 2 Small
Inkling 2 Small
GPT-6.1 Sol
GPT-6.1 Sol
MiniMax M3.1 Flash
MiniMax M3.1 Flash
Ling 3.2 Flash
Ling 3.2 Flash
GPT-6 Luna
GPT-6 Luna
GPT-6 Astra (max)
GPT-6 Astra (max)
Claude Fable 5.2
Claude Fable 5.2
Claude Opus 5.5
Claude Opus 5.5
Claude Fable 5.1
Claude Fable 5.1
Gemini 4 Carbon
Gemini 4 Carbon
GPT-6 Astra (xhigh)
GPT-6 Astra (xhigh)
Kimi K3.1 Pro
Kimi K3.1 Pro
Claude Sonnet 5.5
Claude Sonnet 5.5
Muse Spark 1.3 Max
Muse Spark 1.3 Max
Gemini 4 Argon
Gemini 4 Argon
Grok 4.7
Grok 4.7
Claude Fable 5
Claude Fable 5
Claude Opus 5
Claude Opus 5
Muse Spark 1.3 (xhigh)
Muse Spark 1.3 (xhigh)
Qwen3.8 Max
Qwen3.8 Max
Kimi K3
Kimi K3
Gemini 4 Hydrogen
Gemini 4 Hydrogen
GLM-5.3
GLM-5.3
Grok 4.6
Grok 4.6
GPT-5.6 Sol (xhigh)
GPT-5.6 Sol (xhigh)
Claude Opus 4.8
Claude Opus 4.8
Gemini 3.8 Flash
Gemini 3.8 Flash
Claude Sonnet 5
Claude Sonnet 5
Gemini 3.7 Flash
Gemini 3.7 Flash
Grok 4.5
Grok 4.5
GPT-6 Sol
GPT-6 Sol
Claude Opus 4.6
Claude Opus 4.6
GLM 5.2
GLM 5.2
GPT-6.1 Terra
GPT-6.1 Terra
MiniMax M3
MiniMax M3
Muse Spark 1.2
Muse Spark 1.2
Qwen3.7 Max
Qwen3.7 Max
Gemini 3.6 Flash
Gemini 3.6 Flash
Gemini 3.5 Flash
Gemini 3.5 Flash
Muse Spark 1.1
Muse Spark 1.1
Muse Wonder (max)
Muse Wonder (max)
GLM 5.1
GLM 5.1
GPT-5.6 Terra
GPT-5.6 Terra
Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview
Grok 4.3
Grok 4.3

The Leaderboard

Updated Nov 14, 2026, 10:57 AM UTC

DesignArena by Intelligence
#1
Anthropic logo
?
Claude Opus 5.5 (high)
1447
1704W / 721L
70.3%
±1.8%
2,425
Anthropic
#2
OpenAI logo
?
GPT-6 Astra (max)
1399
1643W / 771L
68.1%
±1.9%
2,414
OpenAI
#3
Anthropic logo
?
Claude Fable 5.2
1371
1157W / 506L
69.6%
±2.2%
1,663
Anthropic
#4
OpenAI logo
?
GPT-6.1 Sol (max)
1369
952W / 860L
52.5%
±2.3%
1,812
OpenAI
#5
Anthropic logo
?
Claude Fable 5.1
1332
2961W / 2008L
59.6%
±1.4%
4,969
Anthropic
#6
Thinking Machines logo
?
Inkling 2 Small
1309
4112W / 3078L
57.2%
±1.1%
7,190
Thinking Machines
#7
Google logo
?
Gemini 4 Carbon
1299
573W / 391L
59.4%
±3.1%
964
Google
#8
Moonshot AI logo
?
Kimi K3.1 Pro
1294
2497W / 1697L
59.5%
±1.5%
4,194
Moonshot AI
#9
Anthropic logo
?
Claude Sonnet 5.5
1287
1650W / 1217L
57.5%
±1.8%
2,867
Anthropic
#10
Meta logo
?
Muse Spark 1.3 Max
1277
7391W / 5562L
57.1%
±0.9%
12,953
Meta
#11
DeepSeek logo
?
DeepSeek-V4.1-Flash (1017)
1276
155W / 165L
48.4%
±5.5%
320
DeepSeek
#12
Google logo
?
Gemini 4 Argon
1271
716W / 882L
44.8%
±2.4%
1,598
Google
#13
xAI logo
?
Grok 4.7
1271
819W / 695L
54.1%
±2.5%
1,514
xAI
#14
Anthropic logo
?
Claude Fable 5
1269
8153W / 5298L
60.6%
±0.8%
13,451
Anthropic
#15
Xiaomi logo
?
MiMo-V2.6-Pro
1265
1899W / 1362L
58.2%
±1.7%
3,261
Xiaomi

Elo Rating: A skill-based rating system starting at 1200. Ratings are calculated using the standard Bradley-Terry model. See more in our Methodology page.Margin of Error: Shows ±standard error from the Bradley-Terry model for Elo rating confidence intervals.