npz

https://github.com/ghostpad/ghostpad

This is a major change behind the scenes, which greatly simplifies the install process - it's now exactly the same as Kobold's install process. npm is no longer required, and the process to run Ghostpad is now exactly the same as the process to run Kobold. Upon installing, you have three UIs to choose from

Ghostpad @ [host]:5000

UI2 @ [host]:5000/new_ui

UI1 @ [host]:5000/classic

Full changelog

- Ghostpad's frontend is now 100% static, built using Bun and Vite.
- Removed "game started" check in chat mode - it will now use chat formatting 100% of the time, even if the story is empty.
- Merged with latest revision from henk's branch
- Changed hostname setting to "Hostname Override".  This will now default to empty, and the socket connection will default to the current URL

• Eliminated many excessive React renders
• Fixed a bug where an action was being updated by index instead of ID, causing broken story updates in some cases
• Added Audio Input and Audio Output options to the top-right menu. With both of these enabled in desktop Chrome, you can now have an audio-only conversation with your AI. These are based on the open Web Speech API spec, but have cross-browser issues and should not be considered a stable feature.

Important: Chrome's implementation of the speech recognition API is not offline - it will send your data remotely.

TLDR: Better performance. less bugs. don't use audio features yet if you dislike Chrome or value privacy

If you are serving Ghostpad from a different host than localhost and want to use speech recognition, you need to enable Insecure origins treated as secure and add that URL in chrome://flags

First post in this brand new community! Welcome.

Ghostpad is an interface built on top of a light fork of KoboldAI. You can check out the project repo for more information.

I'll be sharing updates here as they happen, as well as in the KoboldAI Discord in #ghostpad

This weekend there were a few important backend changes:

1. Support for Exllamav2. After manually installing exllamav2 into the Kobold environment, it should be detected and available in your list of model backends. The reason I'm not auto-installing it is because I've heard this can be quite the project in Windows. My experiences in Linux have been smooth. I hope to make this a more automated process once the build process is simplified.

In Linux, the process should be as simple as cd'ing to your Kobold directory and entering

git clone https://github.com/turboderp/exllamav2.git
sh commandline.sh
pip install ./exllamav2

2. Support for AutoAWQ. If you follow the latest Llama model releases in Huggingface, you've probably noticed a huge number of releases in this format. AutoAWQ is now an auto-installed dependency of koboldai-ghostpad and should not require any additional steps to use. There is a "fuse layers" option which can greatly improve performance, but I've encountered random errors when using it, so I recommend leaving it off for now.

3. Python 3.10 bump. Official KoboldAI is still on Python 3.8, but I'm making the leap so that I can support the type-hinting features used in AutoAWQ. It's possible that this may lead to some unexpected issues, but it's been stable for me so far.

I don't want to make any promises, but one of the features I'm most interested in implementing next is a combination of speech-to-text and text-to-speech, allowing for you to have audio-only conversations with your AI.