Lightning AI records 272,562 monthly visits with -12.3% month-over-month growth and Freemium, Paid, Contact for Pricing pricing. Union.ai records 17,787 monthly visits with -28.4% growth and Freemium, Contact for Pricing pricing.
The AI cloud built by the creators of PyTorch Lightning
The durable AI runtime that runs in your cloud, not ours
| Dimension / Metric | | |
|---|---|---|
| Official Domain | lightning.ai | union.ai |
| Pricing Model | Freemium, Paid, Contact for Pricing | Freemium, Contact for Pricing |
| Monthly Visits | 272.56K (272,562) | 17.79K (17,787) |
| MoM Traffic Delta |
| -38.07K (-12.3%) |
| -7.04K (-28.4%) |
| Primary Market & Share | 🇺🇸 United States (31.8%) | 🇺🇸 United States (52.8%) |
|---|
| Direct Visits / mo | 142.09K | 7.04K |
|---|
| User Rating | 4.7 | 4.4 |
|---|
| Supported Platforms | Website | Website |
|---|
The AI cloud built by the creators of PyTorch Lightning
Spin up single GPUs, full nodes or reserved clusters by the second, selectable from T4 and L4 up to A100, H100, H200, B200 and TPUs, with no long-term contract required.
Ready-to-code environments with CUDA, PyTorch and common libraries preinstalled, usable from the browser, a local IDE or a terminal, with automatic sleep to stop idle spend.
Launch distributed training jobs and large-scale data preparation across multi-node GPU and high-CPU machines, with interruptible instances at up to about 80% off.
Turn a checkpoint into an endpoint or app without rewriting infrastructure code, and host public app links or private model APIs with configurable rate limits.
The durable AI runtime that runs in your cloud, not ours
Tasks, traces and artifacts survive infrastructure failure. Runs fork, recover and replay from the last good step instead of starting over, with retries and caching handled by the runtime.
Named, typed and versioned artifacts wire teams together, with lineage showing which run produced each version and which downstream tasks consume it.
Run the control plane in Union's cloud while workloads execute on clusters in your AWS, GCP, Azure or neocloud account, under your own security perimeter.
Bring OpenAI, Claude, PydanticAI or CrewAI agents; each turn is traced and sandboxed in ephemeral containers, MCP servers and human approvals supported.
SpecialistOptimizes serverless pipelines specifically for generative media models, but lacks general-purpose PyTorch training studios.
The AI developer cloud for GPU pods, serverless inference and clustering
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.
#3Rent cloud GPUs by the second from a global marketplace
Serverless cloud infrastructure for AI and data workloads
An AI-native cloud built from silicon to API for training, tuning, and serving models
SpecialistSimplifies model deployment and fine-tuning via standardized APIs, but offers less granular control over raw GPU clusters.
SpecialistProvides ultra-fast inference for open-weight foundation models, but is less suited for custom distributed model training workflows.
SpecialistLow-cost serverless inference APIs for 200+ open models
Gigawatt-scale AI compute, built and operated end to end
Low-cost cloud GPUs with per-minute instances and per-second sandboxes for AI workloads.
Persistent Ubuntu cloud VMs built for AI agent factories
SpecialistOptimizes serverless pipelines specifically for generative media models, but lacks general-purpose PyTorch training studios.
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.
Rent cloud GPUs by the second from a global marketplace
Serverless cloud infrastructure for AI and data workloads
An AI-native cloud built from silicon to API for training, tuning, and serving models
SpecialistSimplifies model deployment and fine-tuning via standardized APIs, but offers less granular control over raw GPU clusters.
Low-cost cloud GPUs with per-minute instances and per-second sandboxes for AI workloads.
