Displays an animated progress bar below the chat input field that shows how much of the available context window is filled.
Right now, only the llama.cpp, ExLlamav2 and ExLlamav2_HF model loaders are supported.
To use the llama.cpp loader with this extension, you need to apply the following patch in the text-generation-webui folder, which adds two new member variables, tokens_evaluated and tokens_predicted, to the LlamaServer class: https://gist.github.com/mamei16/f7a30f95c4ea1413c0cc49c686342aff