Query alternate LLM from the query entry line

[corsairius]

How to use GitHub

• Please use the 👍 reaction to show that you are interested into the same feature.
• Please don't comment if you have no relevant information to add. It's just extra noise for everyone subscribed to this issue.
• Subscribe to receive notifications on status change and new comments.

---

Feature request

Which Nextcloud Version are you currently using: (see administration page) NC 29.02

Is your feature request related to a problem? Please describe.
Nope.

Describe the solution you'd like
A query for the LLM2 from an alternate LLM in the Docker container was indicated in a post about the Nextcloud LLM2.
Typing:
/model Model_name
would load and use that alternate LLM. Is there such an animal? Similar behavior?

Describe alternatives you've considered
Attempted, but could not really identify anything related to a response specific to the alternate LLM.

Additional context
No

[corsairius]

A Brave Search found the following, but this wasn't reproducible.

Yes, you can select the alternate LLM at the AI prompt in Nextcloud.

To do so, follow these steps:

Open the AI Prompt: Open the AI prompt in Nextcloud where you would normally interact with the default LLM.
Use the /model Command: Type /model followed by the name of the alternate LLM you want to use. For example: /model llama2
Select the Model: If the model is available, you will be able to select it from the dropdown list or by typing its name.
Confirm the Model Change: Once you select the alternate LLM, confirm the change by typing yes or selecting the Confirm button.
After selecting the alternate LLM, the AI prompt will use the newly selected model for all subsequent interactions until you switch back to the default model or select a different one.

Note: The availability of this feature may depend on the specific version of the llm2 app and Nextcloud you are using. Make sure to check the documentation for your specific setup to confirm the functionality.

[marcelklehr]

Hi @corsairius
That's an interesting idea, thank you for the feedback!

[corsairius]

You're welcome! It was something I would do at different times. I'm looking for legal vs technical vs health LLMs, and found that post above. I wasn't hurting for it, just looking for confirmation as to whether that works or not. I couldn't see a difference because the LLM I'm using from the Health Ranger (Neo) and soon (Epoch) only covers the health aspect of LLMs, and switching to the more public version and back wasn't repeatable in context. If the LLM isn't targeted to a specific area, the results are more random, and not domain specific. So targeting a specific knowledge domain (family server) was more the goal.

Thanks!

[tuxArg]

Yes, you can select the alternate LLM at the AI prompt in Nextcloud.

I can't see it working for me. Is it implemented already?
It would be nice to add a temperature parameter too.

[corsairius]

Yes, you can select the alternate LLM at the AI prompt in Nextcloud.

I can't see it working for me. Is it implemented already? It would be nice to add a temperature parameter too.

That was a post generated from the Brave ML engine. It is not implemented at this time afaik.