Closed CT1800098 closed 1 year ago
What do you want to work on? Backend Inference or Frontend
what is the differece between them?
What do you want to work on? Backend Inference or Frontend
I sort of want to do what the web version does
The docs detail how you can setup and start backend on your local PC
Create a venv Clone the repo and install the required libraries as guided in the backend readme file Make sure you have postgres and redis running Start the backend server using the script inside the backend directory ( Again, instructions in the readme file of the backend directory)
Check documentation in your local machine at localhost:8080/docs
The docs detail how you can setup and start backend on your local PC
Create a venv Clone the repo and install the required libraries as guided in the backend readme file Make sure you have postgres and redis running Start the backend server using the script inside the backend directory ( Again, instructions in the readme file of the backend directory)
Check documentation in your local machine at localhost:8080/docs So I just follow the instruction on the redme in the backend?
A small remark: unless you have a very powerful GPU you will be only able to run the website locally, but not the chat (=Inference).
what is the differece between them?
Backend is to work on the open assistant web backend Front is to improve the UI for the same web Inference is to work on the LLMs
A small remark: unless you have a very powerful GPU you will be only able to run the website locally, but not the chat (=Inference).
how many VRM do I need?
what is the differece between them?
Backend is to work on the open assistant web backend Front is to improve the UI for the same web Inference is to work on the LLMs
what is LLM?
what is the differece between them?
Backend is to work on the open assistant web backend Front is to improve the UI for the same web Inference is to work on the LLMs
what is LLM?
The chat itself Large language models They have so many parameters you need more hardware like @stefangrotz said
what is the differece between them?
Backend is to work on the open assistant web backend Front is to improve the UI for the same web Inference is to work on the LLMs
what is LLM?
The chat itself Large language models They have so many parameters you need more hardware like @stefangrotz said
how many VRM do I need to run?
For the 70b model you'll need 48GB VRAM.
If you want to run models locally I recommend https://gpt4all.io/ Its for language models that are optimized for consumer hardware and it is easy to use.
For the 70b model you'll need 48GB VRAM.
If you want to run models locally I recommend https://gpt4all.io/ Its for language models that are optimized for consumer hardware.
is it censored or uncensored?
I have no idea on what should I do to get the local version up and running after install it on my pc