Closed 0x000011b closed 1 year ago
What it says on the tin. We use Kobold as an inference server so we can fit a 6B model on a Colab GPU (16GB T4s as of the time I'm writing this), and offer a notebook to automatically start it up and launch the Gradio UI to talk to it.
What it says on the tin. We use Kobold as an inference server so we can fit a 6B model on a Colab GPU (16GB T4s as of the time I'm writing this), and offer a notebook to automatically start it up and launch the Gradio UI to talk to it.