Hi All, I am new to setting up self hosted version of n8n. I use docker’s container to host the n8n, but I found that deepseek-r1:8b and gemma3:4b do not support tool. If I want to use tool in my local n8n, what should I do? Thanks in advance!
deepseek-r1:8b
gemma3:4b
Yes, you have to download better models :
Qwen 3 (MoE model, supports tool calling)
Granite 3.3
Mistral-small3.1
phi4-mini
etc…
Tool calling isn’t available without extra work for small models.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
