Atlas

Benchmarks

← All benchmarks

Arena Code WebDev — Content Creation Tools

Code

Arena Code WebDev rating for content-creation and editing tools. Anonymous generated applications are compared through blind pairwise user votes.

Top models (higher is better)

ModelScore
Claude Opus 51678
Kimi K31641
Claude Fable 51594
Opus 4.81562
Opus 4.71551
GLM-5.21543
Opus 4.61542
Gemini 3.6 Flash1532
Muse Spark 1.11529
Grok 4.51521
Hy31516
Sonnet 4.61508
GLM-5.11503
Sonnet 51501
Gemini 3.5 Flash1494
Kimi K2.61484
MiniMax M31483
MiMo-V2.5-Pro1472
Qwen3.6 Plus (2026-04-02)1453
Kimi K2.7 Code1452
GPT-5.3-Codex1449
Opus 4.51446
MiniMax M2.51435
DeepSeek-V4-Pro1432
Gemini 3.1 Pro Preview1431
MiMo-V2.51430
MiMo-V2-Pro1430
GLM-51427
Kimi K2.51426
GLM-4.71410
Gemini 3 Flash Preview1404
MiniMax M2.71402
GPT-5.21402
Gemini 3 Pro Preview1397
GPT-51397
Qwen3.5 397B A17B1397
GPT-5.4 Mini1394
GPT-5.11390
GPT-5.41388
Sonnet 4.51384
Loading Atlas data…