Code Arena 每月有 35,827,928 次造訪,月增率 +19.4%,定價模式為免費。Vast.ai 每月有 1,257,554 次造訪,成長率 +2.8%,定價模式為付費。
| 面向/指標 | ![]() | |
|---|---|---|
| 官方網域 | arena.ai | vast.ai |
| 定價模式 | 免費 | 付費 |
| 每月造訪次數 | 3582.79萬 (35,827,928) | 125.76萬 (1,257,554) |
| 月流量變化量 | +583.28萬 (+19.4%) | +3.42萬 (+2.8%) |
| 主要市場與市佔 | 🇮🇳 印度 (13.1%) | 🇺🇸 美國 (16.7%) |
| 每月直接造訪 | 2433.99萬 | 93.71萬 |
| 使用者評分 | 4.7 | 4.6 |
| 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.
從全球市集以秒為單位租用雲端 GPU
從超過 20,000 顆 GPU、60 多種型號(從 RTX 3060 等級到 GB200)的全球池中租用,橫跨 40 多座資料中心。
可選保證開機時間的隨需、便宜 50% 以上的可中斷,或最高 50% 折扣並含容量保證的預留方案。
以秒計費、無最低時數,只用你實際使用的算力,讓實驗與短時工作也能維持低成本。
透過 CLI、Python SDK 或 REST API 完全以程式碼佈建與最佳化執行個體,這也讓平台適合代理程式使用。

專項更強Together AI is the AI Native Cloud — a full-stack platform for running, fine-tuning and training open-source models. It spans serverless inference APIs, dedicated endpoints, GPU clusters, post-training and storage, all backed by in-house systems research.
專項更強Together AI is the AI Native Cloud — a full-stack platform for running, fine-tuning and training open-source models. It spans serverless inference APIs, dedicated endpoints, GPU clusters, post-training and storage, all backed by in-house systems research.
Provides enterprise-grade AI cloud infrastructure with managed ML tools, replacing peer-to-peer rentals with reliable data centers.
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
免費 / 開源檢索並研究數百萬公開的 SKILL.md 代理技能

Builds large-scale data center GPU clusters tailored for heavy model training, lacking peer-to-peer spot pricing.
Builds large-scale data center GPU clusters tailored for heavy model training, lacking peer-to-peer spot pricing.
面向 Cursor 的社群外掛、規則與 MCP 伺服器中心