v0.0.3

• This version comes with performance improvements for arm64 devices by compiling llama.cpp with Arm64-specific CPU
  flags. (#18)
• The chat messages can now highlight links/URLs and make them clickable.
• The height of the list in the select model dialog is fixed to avoid overlapping with the Add Model button. (#12)
• The dialog to select models includes a button to sort models by name or date added.
• The time taken (in seconds) to generate the response is now displayed below the last response from the LLM. (#7)

v0.0.2

• Clear previous chat messages when LLMInference::load_model is called
• Allow rendering non-ASCII characters on the chat interface generated by the LLMs/SLMs
• Show token generation speed (in tokens/second) for the latest message in the chat interface (#1)

v0.0.1

This release:

• adds a button to stop response generation
• disables chat options when no chat is selected