CLI Profile

Ollama CLI

Community CuratedOllamavlatest

Run local LLM models and inference workflows.

Install Matrix

brew

hash: unavailable

Auth Guide

None

Token refresh: No auth required for local model runtime.

Step 1

No explicit authentication required for local usage.

Command Map

Global Flags

--help (string|boolean) - help for serve
--version (string|boolean) - Show version information

Ai Inference

ollama serve

Start ollama

Output: Tableai-inferencedeveloper-loop

Example 1

Example 2

Known errors: Model access denied; Inference job failed; Provider quota exceeded

ollama create

Create a model

Output: Plain textai-inferencedeveloper-loop

Example 1

Example 2

Known errors: Invalid flag or payload for requested operation; Inference job failed; Provider quota exceeded

ollama show

Show information for a model

Output: JSONai-inferencedeveloper-loop

--licenseoptional (string|boolean) - Show license of a model
--modelfileoptional (string|boolean) - Show Modelfile of a model
--parametersoptional (string|boolean) - Show parameters of a model
--systemoptional (string|boolean) - Show system message of a model

Example 1

Example 2

Known errors: Resource not found for current context; Inference job failed; Provider quota exceeded

ollama run

Run a model

Output: Tableai-inferencedeveloper-loop

--hidethinkingoptional (string|boolean) - Hide thinking output (if provided)
--insecureoptional (string|boolean) - Use an insecure registry
--nowordwrapoptional (string|boolean) - Don't wrap words to the next line automatically
--thinkoptional (string|boolean) - Whether to use thinking mode for supported models

Example 1

Example 2

Known errors: Model access denied; Inference job failed; Provider quota exceeded

ollama stop

Stop a running model

Output: Tableai-inferencedeveloper-loop

Example 1

Example 2

Known errors: Model access denied; Inference job failed; Provider quota exceeded

ollama pull

Pull a model from a registry

Output: Tableai-inferencedeveloper-loop

--insecureoptional (string|boolean) - Use an insecure registry

Example 1

Example 2

Known errors: Model access denied; Inference job failed; Provider quota exceeded

ollama push

Push a model to a registry

Output: Plain textai-inferencedeveloper-loop

--insecureoptional (string|boolean) - Use an insecure registry

Example 1

Example 2

Known errors: Model access denied; Project or environment not linked; Build/deploy operation failed on remote provider

Workflow Recipes

AI Inference

Authenticate model providers and execute inference runs.

1ollama

→

2huggingface-cli

→

3replicate

Open workflow detail

CLI Profile

Loading CLI profile...

Fetching install matrix, auth flow, command map, and compatibility data.

Please wait while clime prepares the listing details.