Use of tensorzero in N8N #5361
Replies: 1 comment 6 replies
-
|
Glad to hear! I haven't tried n8n + TensorZero, but it looks like they support OpenAI-compatible models, so it should work. You should set the base URL to your TensorZero base URL (e.g. |
Beta Was this translation helpful? Give feedback.
-
|
@GabrielBianconi After some tests I found out that when declaring the credentials in n8n the logs in tensorzero show this: |
Beta Was this translation helpful? Give feedback.
-
|
@MacherelR We support chat completions and embeddings. We also support a translation layer between chat completions and responses. We don't have a models endpoint. This is tracked here: #5032 |
Beta Was this translation helpful? Give feedback.
-
|
I see. Then I guess for that use case it might be needed to keep working with the raw http request module and not the openai as it seems to use the |
Beta Was this translation helpful? Give feedback.
-
|
@GabrielBianconi Any insight or ETA on the feature or isn't it planned at all ? |
Beta Was this translation helpful? Give feedback.
-
|
@MacherelR We don't have an ETA. I do see the value in this feature so we might build it at some point, but the team is focused on other features right now. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I have successfully deployed tensorzero, and interfaced it with ollama for models inference. It works amazingly and I for example set it up with
sgpt(s.o https://github.com/tbckr/sgpt).Now someone in my organisation asked me how to use it inside n8n, and I couldn't find another way than http requests directly, which is not a great UX. Has anyone ever tried it and had success in integrating tensorzero in n8n ?
Beta Was this translation helpful? Give feedback.
All reactions