Atlas

Benchmarks

← All benchmarks

Arena Code WebDev — Data & Analytics

Code

Arena Code WebDev rating for data and analytics applications. Anonymous generated applications are compared through blind pairwise user votes.

Top models (higher is better)

ModelScore
Claude Opus 51625
Kimi K31595
Claude Fable 51556
GLM-5.21544
Muse Spark 1.11534
Opus 4.71530
DeepSeek-V4-Flash1527
Opus 4.81526
Sonnet 4.61517
Opus 4.61517
Grok 4.51514
GLM-5.11504
Sonnet 51496
Gemini 3.6 Flash1494
Kimi K2.61490
MiniMax M31484
Hy31482
Gemini 3.5 Flash1471
MiMo-V2.5-Pro1464
Kimi K2.7 Code1453
Qwen3.6 Plus (2026-04-02)1450
Opus 4.51447
MiMo-V2.51443
DeepSeek-V4-Pro1436
MiMo-V2-Pro1433
GLM-51427
GPT-5.21422
Kimi K2.51422
GLM-4.71420
Gemini 3 Pro Preview1419
Gemini 3 Flash Preview1412
GPT-5.41408
Gemini 3.1 Pro Preview1405
Inkling1402
MiniMax M2.71402
GPT-5.3-Codex1399
Qwen3.5 397B A17B1395
GPT-5.4 Mini1389
GPT-5.11388
MiniMax M2.51387
Loading Atlas data…