v1.1.1 Firefox support, bugfixes

BrowserLlama Setup Instructions

To set up the software, follow these steps:

1. Download the appropriate package:

• For CPU (no CUDA): Download browserllama.zip, which includes a koboldcpp binary that runs on CPU.
• For GPU (with CUDA): Download browserllama.gpu.inference.zip, which includes the CUDA version (note: this file is much larger).
• To use your own koboldcpp binary: Download browserllama.no.kobold.zip. After extraction, copy your own binary into the host folder and name it koboldcpp.exe.

2. Extract the downloaded folder.

3. Run the install.bat file to install the software.

4. Language Model:

You will also need a model in the .gguf format. Download a recommended smaller model from here.

---

This version of the backend software is compatible with the latest extension (v1.1) and may support future versions as well.