Version 4.3

• Redesigned the chat history
• Renamed the profile names of vicuna-v0 and vicuna-v1.1
• Updated the README

Version 4.2

• Bumped llama-cpp-python[server] from 0.1.54 to 0.1.56
• Changed the profile format
• Added support for Vicuna v1.1
• Added support for Manticore Chat

Version 4.1

• Bugfix: Preserve whitespaces in messages #10
• minor code improvements

Version 4.0

BREAKING! Bumped llama-cpp-python[server] from 0.1.50 to 0.1.54

This (again) requires re-quantized models. The new format is ggml v3. See ggerganov/llama.cpp#1508

Version 3.1

• Added a toggle button for the sidebar
• Implemented a responsive design (fixes #4)
• Made more minor improvements to the frontend

Version 3.0

Changed the frontend to support other LLMs and added support for Vicuna.

Version 2.1

Added more parameters to the sidebar: top_k, repeat_penalty, presence_penalty, frequency_penalty

Version 2.0

(Breaking) Bumped llama-cpp-python[server] from 0.1.48 to 0.1.50
You may have to re-quantize your models, see ggerganov/llama.cpp#1405

Version 1.1

v1.1

Minor style improvements and fixes

Version 1.0

v1.0

Pinned all pip dependencies in requirements.txt and added dependabot …