Code Arena: 提供並排基準測試與社群排行榜，但專門評估程式設計與程式碼生成模型。 Code Arena 每月有 35,827,928 次造訪，月增率 +19.4%，定價模式為免費。Design Arena 每月有 949,534 次造訪，成長率 -13.2%，定價模式為免費。
Code Arena is a hands-on playground and leaderboard where you can pit leading AI coding models against each other on real web builds, comparing speed, accuracy, and logic in live side-by-side battles.
讓 AI 模型同台競技、由真人投票選出最佳設計產出的群眾外包競技場
| 面向／指標 | | |
|---|---|---|
| 官方網域 | arena.ai | designarena.ai |
| 定價模式 | 免費 | 免費 |
| 每月造訪次數 | 3582.79萬 (35,827,928) | 94.95萬 (949,534) |
| 月流量變化量 | +583.28萬 (+19.4%) | -14.4萬 (-13.2%) |
| 主要市場與市佔 | 🇮🇳 印度 (13.1%) | 🇮🇳 印度 (24.2%) |
| 每月直接造訪 | 2433.99萬 | 56.46萬 |
| 使用者評分 | 4.7 | 4.4 |
| 支援平台 | 網站 | 網站 |
Code Arena is a hands-on playground and leaderboard where you can pit leading AI coding models against each other on real web builds, comparing speed, accuracy, and logic in live side-by-side battles.
Send the same coding prompt to multiple leading models and compare their generated apps and logic side by side.
Watch models construct web apps and websites live, so you can see process and output, not just the final file.
Projects go beyond one-off snippets, testing how well a model structures components, routes, and styles together.
Track which coding models accumulate the most wins over time to identify consistently strong performers.
讓 AI 模型同台競技、由真人投票選出最佳設計產出的群眾外包競技場
輸入提示詞，查看多個主流模型的回答，再把票投給最好的產出——這是讓評測保持群眾外包與獨立的核心機制。
公開排名涵蓋網站與前端建置、圖像與圖像編輯、影片與圖生影片、3D、SVG、簡報、試算表、PDF、程式碼藝術與語音合成。
結果以 Elo/Bradley-Terry 風格的評分機制彙整，因此排名反映的是集體人類審美，而非廠商自報的分數。
為代理式編碼、Godot 遊戲開發、React Native、Android（Kotlin）與全端建置提供專門排行榜，協助開發者選出真正適合工作的模型。
免費 / 開源檢索並研究數百萬公開的 SKILL.md 代理技能
面向 Cursor 的社群外掛、規則與 MCP 伺服器中心
專項更強Midlibrary is an independent, non-official catalog of Midjourney artistic styles, SREF codes, prompts and workflow guides. Search by look, artist category or keyword, copy the prompt or style reference code, and paste it into Midjourney to reproduce the aesthetic in your own generations.
專項更強Midlibrary is an independent, non-official catalog of Midjourney artistic styles, SREF codes, prompts and workflow guides. Search by look, artist category or keyword, copy the prompt or style reference code, and paste it into Midjourney to reproduce the aesthetic in your own generations.
專項更強提供公開的模型基準測試與效能排行榜，但專注於電腦操作與行動代理而非設計。
專項更強會隨著時間記錄相同提示詞下的模型歷史輸出結果，但缺乏即時群眾投票與互動排名功能。
專項更強一站比較各大廠商 LLM API 權杖價格
專項更強在策略博弈中檢驗 AI 的合作、談判與欺騙行為
Helps users compare model performance and AI stacks, but offers less specialized benchmarking for visual design tasks.
