Ollama Cloud Setup

Ollama cloud setup is useful when you want an Ollama-compatible provider but do not want to run the model on your own laptop or workstation. This can be a good hybrid option: the configuration feels similar to Ollama local, but inference happens on a remote endpoint.

When to use Ollama cloud

• Your local machine does not have enough memory or GPU capacity.
• You want an Ollama-compatible API but prefer remote infrastructure.
• You manage a shared model endpoint for a team.
• Your organization has an approved private LLM endpoint.

Configuration checklist

1. Open MLJAR Studio provider settings.
2. Select Ollama Cloud as the provider.
3. Enter your Ollama API key.
4. Enter the model name, for example qwen3.5:397b or gemma4:31b.
5. Click Test connection to check whether the API key works and the model is available.
6. When the test succeeds, click Save provider.
7. Confirm the success toast.
8. Check the top provider chip in the sidebar. It should show Ollama Cloud with a green dot.

Example model names

The model name depends on the models available in your Ollama Cloud account.

qwen3.5:397b
gemma4:31b

Privacy note

Ollama cloud is not the same as local Ollama. With local Ollama, the model runs on your own computer. With a remote endpoint, prompts and context are sent to that remote endpoint. Use this only when the endpoint matches your privacy and security requirements.

Related pages

If you want a local-only workflow, use Ollama Local Setup. If you are choosing between local and cloud providers, read Local vs Cloud LLMs.