Hello everyone.
Could someone advise on how to modify the WebUI interface to ensure that combined selection lists are available/accessible to screenreaders lika NVDA or Jaws? Previously, in older versions of the Text Generation Web UI, I was able to toggle between loaders such as LlamaCpp and Transformers via the keyboard, which is no longer functional. Furthermore, I am unable to locate the dedicated tab for model training.
My setup involves a low-spec laptop with 8GB of RAM and 4GB of VRAM, yet I can successfully load smaller 4B models. At present, my goal is to experiment with model training, but the lack of accessibility for screenreaders makes this impossible."
Hello everyone.
Could someone advise on how to modify the WebUI interface to ensure that combined selection lists are available/accessible to screenreaders lika NVDA or Jaws? Previously, in older versions of the Text Generation Web UI, I was able to toggle between loaders such as LlamaCpp and Transformers via the keyboard, which is no longer functional. Furthermore, I am unable to locate the dedicated tab for model training.
My setup involves a low-spec laptop with 8GB of RAM and 4GB of VRAM, yet I can successfully load smaller 4B models. At present, my goal is to experiment with model training, but the lack of accessibility for screenreaders makes this impossible."