Skip to content
BENIBIX Learn

Cloud AI Platforms

Managed AI services from the big clouds · 23 terms

Sort
Compare

Cloud AI Basics

8

The ideas every big cloud uses for its AI: how you pay, and where your data lives.

AI models and tools you use from inside your own cloud account.

A cloud runs the model for you, and you pay for each use.

You rent a fixed share of a model, for a fixed price.

A cap on how many tokens an app may use each minute. Over it, you wait.

Your data, and the AI that reads it, stay in one place you pick.

A promise from Microsoft. Its EU cloud data stays in Europe.

A promise that an AI firm keeps no copy of what you send or get back.

A ready-made cloud tool for one AI job. It can turn speech into text or translate.

GPU Compute

5

Renting GPUs to train and run models, by the hour or only while requests come in.

A service that rents out computers with GPUs by the hour. Teams use them for AI.

A cloud firm that rents out big groups of NVIDIA GPUs. AI labs train on them.

A GPU cloud for coders. Rent a GPU machine, or use one only when called.

GPUs that switch on when a request comes. They drop to zero when no one asks.

Cloudflare runs AI models on its GPUs around the world. You pay as you go.

Model Platforms

7

The cloud services that host models from many makers and help you build on them.

One AWS tool that reaches AI models from many makers.

It runs your AI agents on AWS, with memory and tools.

Where you build AI apps and agents on Azure.

The same GPT models, but run in Azure, not by their maker.

Google Cloud's home for AI models and agents. It used to be called Vertex AI.

A free Google site to try Gemini and get a key.

IBM's place to build, tune and run AI models, in the cloud or on site.

ML Platforms

3

Cloud tools for training your own models, from notebooks to full setups.

The AWS tool for training and running your own models.

Azure's cloud tool to train, launch and run your own ML models.

Free notebooks on the web that can borrow a GPU.