RunPod 每月有 2,881,043 次造訪，月增率 +5.9%，定價模式為付費。Thunder Compute 每月有 102,147 次造訪，成長率 -29%，定價模式為 Paid (per-minute instances, per-second sandboxes)。
低成本雲端 GPU：面向 AI 工作負載的按分鐘執行個體與按秒計費沙箱。
| 面向／指標 | | |
|---|---|---|
| 官方網域 | runpod.io | thundercompute.com |
| 定價模式 | 付費 | Paid (per-minute instances, per-second sandboxes) |
| 每月造訪次數 | 288.1萬 (2,881,043) | 10.21萬 (102,147) |
| 月流量變化量 | +16.15萬 (+5.9%) | -4.17萬 (-29%) |
| 主要市場與市佔 | 🇺🇸 美國 (26.1%) | 🇺🇸 美國 (53.5%) |
| 每月直接造訪 | 163.48萬 | 4.07萬 |
| 使用者評分 | 4.7 | 4.4 |
| 支援平台 | 網站 | 網站 |
面向 AI 開發者的 GPU Pod、Serverless 推論與叢集雲平台
按需、可中斷或預留方式租用 GPU 執行個體，涵蓋 30 多種 GPU 型號，從 RTX 4090 到 B200、B300 等級加速卡，部署於全球 31 個區域。
把任意容器部署為 API 端點，按需從 0 擴展到數千個 worker，僅為實際使用的算力付費。
啟動多節點 GPU 叢集用於分散式訓練與大規模推論，可掛載共享儲存，且無需長期承諾。
從精選中樞部署熱門開源模型，或使用公開端點呼叫圖像、音訊與文字模型，無需自建基礎設施。
低成本雲端 GPU：面向 AI 工作負載的按分鐘執行個體與按秒計費沙箱。
開通 RTX A6000、L40、A100 80GB 或 H100 SXM 機器，通常幾分鐘內即可 SSH 登入，僅在執行時計費。
啟動帶或不帶 GPU 的隔離 Ubuntu 虛擬機，用於 RL rollout、評估與 CI，按秒計費，工作之間可縮容至零。
CLI、VS Code 擴充套件、MCP 伺服器與 REST API，讓你能從終端機、編輯器、AI Agent 或自有服務驅動基礎設施。
將執行個體磁碟保存為快照並在之後還原，實驗可隨時暫停，無需讓機器持續執行。
專項更強Provides a unified API gateway across hundreds of models, but entirely abstracts away underlying GPU infrastructure management.
專項更強Provides a unified API gateway across hundreds of models, but entirely abstracts away underlying GPU infrastructure management.
專項更強Specializes in lightning-fast serverless execution for generative media models, but lacks general-purpose GPU pod rental options.
Nebius is an AI-focused cloud platform that gives teams non-virtualized NVIDIA GPU clusters, managed MLOps tooling, and a serverless inference layer so they can move from experiment to production on one stack.
Offers a polished API and cloud platform for running open models, but provides less direct infrastructure-level control than RunPod.
Offers a polished API and cloud platform for running open models, but provides less direct infrastructure-level control than RunPod.
免費 / 開源由 PyTorch Lightning 創建團隊打造的 AI 雲端平台
FluidStack builds and operates large-scale compute infrastructure for AI, including gigawatt-class data centers and GPU clusters used for training and inference. The company designs its own power, cooling and data center systems, onshores manufacturing and deploys solar generation to supply the power its clusters need. Rather than a self-serve cloud console, FluidStack works with AI labs and enterprises to provision large reserved clusters, with orchestration tooling for the ML frameworks and schedulers those teams already use.
FluidStack builds and operates large-scale compute infrastructure for AI, including gigawatt-class data centers and GPU clusters used for training and inference. The company designs its own power, cooling and data center systems, onshores manufacturing and deploys solar generation to supply the power its clusters need. Rather than a self-serve cloud console, FluidStack works with AI labs and enterprises to provision large reserved clusters, with orchestration tooling for the ML frameworks and schedulers those teams already use.
