Using HostedGPT with a Local LLM like Gemma 2 9B #471
krschacht
announced in
Announcements
Replies: 1 comment 2 replies
-
It works great! If the same model is provided by multiple API services, the user could be confused about which one to choose from the list in the new assistant screen. Maybe adding a filter for "API services" on that screen or showing the API service in the list (e.g. Gemma 2 9b - LM Studio) could avoid this ambiguity. |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
We recently added the ability for HostedGPT to point to any API server, including local servers. You can now run the app completely free if you download a model to your computer.
I've been trying out the new Gemma 2 models. My computer is not powerful enough to run the 27B model so I'm using Gemma 2 9B. Here are the steps, in case anyone would like to try it.
I recommend LM Studio but a lot of other people use Llama.cpp / Llamafiles so those should work too.
LM Studio
openai
http://localhost:1234/
Gemma 2 9b
bartowski/gemma-2-9b-it-GGUF
LM Studio
Gemma 2 9b
Gemma 2 9b
If you try this, I'm eager to hear how well it works — both the setup and the model itself.
Beta Was this translation helpful? Give feedback.
All reactions