Unable to setup Ollama credential

[weilies_chok]

Describe the problem/error/question

Unable to authenticate Ollama (screenshot shared below)

Please share your workflow

Share the output returned by the last node

Information on your n8n setup

• n8n version: 1.58.2
• Database (default: SQLite): default
• n8n EXECUTIONS_PROCESS setting (default: own, main): default
• Running n8n via (Docker, npm, n8n cloud, desktop app): docker
• Operating system: window 11

i have my olllama running in my localhost (as screenshot below), but no luck with the credential

But based on doc, there is an issue if i self host n8n.

Ollama credentials | n8n Docs

here is my docker, there isn’t any ollama

And i still no clue how to setup variable in n8n…Can any guru give me some pointers? thanks~

[ria]

Hi @weilies_chok

Since you’re hosting n8n in docker you will need to change the Ollama credentials to http://host.docker.internal:11434/ as the host.

Have a look here:

GitHub - n8n-io/self-hosted-ai-starter-kit: The Self-hosted AI Starter Kit is an open-source...

Hope it helps!

[weilies_chok]

it’s working now ria, thanks!

but here come another question, it’s using the latest llama3.2, which i guess should be 3b

is there a way to downgrade to 1b? coz i m using an old laptop, the response of the chat is very slow

[ria]

Hi @weilies_chok

That’s great to hear!

If you’ve downloaded the 1b model version you should be able to select that from the Model drop-down

[weilies_chok]

that’s the thing, coz the “self hosted ai kit” GitHub - n8n-io/self-hosted-ai-starter-kit: The Self-hosted AI Starter Kit is an open-source template that quickly sets up a local AI environment. Curated by n8n, it provides essential tools for creating secure, self-hosted AI workflows., bundle everything (n8n, ollama (only 1 model), postgrest…)

and days ago, i install my own, and i hv the model, but it seems doesn’t link with n8n