Ollama logo
ConnectorOllama

Ollama integration for Claude and Codex

Create a Ollama connection, then control which Spaces can use its approved tools and data without pasting credentials into prompts or threads.

Use Ollama from Claude

Add the Ollama connection to a Space that runs Claude, then use it from every channel in that Space.

Use Ollama from Codex

Add the Ollama connection to a Space that runs Codex, then use it from every channel in that Space.

Use Ollama from Sidekick

Connect your own OpenClaw to Sidekick while type.com provides the conversations, permissions, connections, skills, and automations.

Developer Tools & IT

What the Ollama connector can do

Ollama lets a team run open language models on its own machines or cloud servers, instead of relying only on hosted APIs. With an Ollama connection in type.com, a development Space can ask Claude to chat with a local model, generate raw text output, or list which models are currently installed, useful when comparing results against Claude itself.

One connection, many Spaces

Connect Ollama once, then decide which Spaces can use it in threads, skills, automations, and coding work.

What teams do with Ollama in type.com

Compare a local model's answer

Send the same question to an Ollama model side by side with Claude's own answer to compare outputs.

Try asking
Chat with our Ollama llama3 model about this support question and show me its answer.

Check which models are installed

List the models currently available on the connected Ollama instance before deciding which one to use for a task.

Try asking
List the Ollama models installed on our server right now.

Generate raw text without templating

Use raw mode to send a fully custom prompt to Ollama when full control over formatting matters more than convenience.

Try asking
Generate text with Ollama in raw mode using this exact prompt, no templating.

Representative actions

  • Chat with Ollama model

    Sends a chat message with conversation history to an Ollama model for a multi-turn exchange.

  • Generate Text with Ollama

    Generates a text response from an Ollama model, with an optional raw mode that skips prompt templating.

  • List Models

    Lists the models installed on the connected Ollama instance, with size and modification details.

  • OpenAI-Compatible Chat Completion

    Creates a chat completion from an Ollama model using the OpenAI API response format.

  • OpenAI-Compatible Text Completion

    Generates a text completion from an Ollama model in OpenAI-compatible format, beyond chat-style requests.

Connection

API and auth details

Ollama documents local and cloud-compatible APIs for running large language models. Integrations can generate completions, chat with models, create embeddings, list and pull models, manage model files, inspect running models, and connect local model inference to agent or application workflows.

Source
type.com catalog metadata
Auth schemes
API key
Official site
https://ollama.com/
FAQ

Questions people ask before connecting Ollama

Can Claude talk to local Ollama models from type.com?

Yes. Once an Ollama connection is shared with a Space, the agent can chat with a model, generate text, or list installed models from a thread.

How do I authenticate Ollama in type.com?

Ollama uses an API key, added once by an admin when the connection is created.

Can Ollama comparisons run on a schedule?

Yes. A recurring comparison, such as testing a model weekly against new prompts, can be set up as a type.com automation.

Is this the same as a Ollama MCP server?

Not exactly. type.com uses connectors and connections to give selected Spaces access to approved app tools and data. Some connectors use hosted MCP, while others use OAuth, API keys, service accounts, or custom APIs.

More developer tools & it connectors

All Developer Tools & IT integrations