Thunder Compute 记录了 102,147 月访问量，月度环比增长为 -29%，定价为 Paid (per-minute instances, per-second sandboxes)。Union.ai 记录了 17,787 月访问量，增长率为 -28.4%，定价为免费增值，定价请咨询。
低成本云 GPU：面向 AI 工作负载的按分钟实例与按秒计费沙箱。
| 维度 / 指标 | |  |
|---|---|---|
| 官方域名 | thundercompute.com | union.ai |
| 定价模式 | Paid (per-minute instances, per-second sandboxes) | 免费增值，定价请咨询 |
| 月访问量 | 10.21万 (102,147) | 1.78万 (17,787) |
| 月度环比流量变化 | -4.17万 (-29%) | -7039 (-28.4%) |
| 主要市场与份额 | 🇺🇸 美国 (53.5%) | 🇺🇸 美国 (52.8%) |
| 直接访问 / 月 | 4.07万 | 7040 |
| 用户评分 | 4.4 | 4.4 |
| 支持的平台 | 网站 | 网站 |
低成本云 GPU：面向 AI 工作负载的按分钟实例与按秒计费沙箱。
开通 RTX A6000、L40、A100 80GB 或 H100 SXM 机器，通常几分钟内即可 SSH 登录，只在运行时计费。
启动带或不带 GPU 的隔离 Ubuntu 虚拟机，用于 RL rollout、评估与 CI，按秒计费，任务之间可缩容至零。
CLI、VS Code 扩展、MCP 服务器与 REST API，让你能从终端、编辑器、AI Agent 或自有服务驱动基础设施。
将实例磁盘保存为快照并在之后恢复，实验可随时暂停，无需让机器持续运行。
在你自己的云中运行的持久化 AI 运行时
任务、追踪与制品可跨基础设施故障存活。运行可派生、恢复并从最后一个成功步骤重放，重试与缓存由运行时统一处理。
具名、带类型并带版本的制品把团队连接起来，血缘图清晰展示某个版本由哪次运行产出、又被哪些下游任务消费。
控制面托管在 Union 云，负载在你自己的 AWS、GCP、Azure 或新云集群上执行，始终处于自有安全边界内。
支持接入 OpenAI、Claude、PydanticAI、CrewAI 等智能体，每一轮都被追踪并在临时容器中沙箱执行，支持 MCP 服务器与人工审批。
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.
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.
从全球市场按秒租用云 GPU
面向 AI 与数据负载的无服务器云基础设施
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.
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.
免费 / 开源编排全球闲置消费级 GPU 的分布式 GPU 云
由 PyTorch Lightning 创建团队打造的 AI 云平台
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.
面向训练与推理工作负载的私有 AI 云
