v2.3.0

Update the default API to Qwen3

Enable reasoning effort support

v2.2.0

Update Default API.

Layout Loading Capability.

DeepSeek-R1 and Reasoning Content Support.

v2.1.0

Support multiple prompts and visualization of MCP.

Update the icon, tooltip and interaction logic of the input box.

v2.0.0

Support MCP (Model Context Protocol) by acting as a renderer

Enhance specific chat card functionalities, such as copy, delete, source text display, and bulk tool call deletion

Update the default LLM to the latest Llama-3.3

v1.5.0

Support i18n
Add HuggingFace demo

v1.4.0

Support interruption and re-generation
Support multimodal vision model
Support display toggling between original format and Markdown format

v1.3.0

Update new models
Support export of chat history
Add template for Nvidia NIM

v1.2.0

Support new model and cloudflare AI API

v1.1.0

Update description

v1.0.0

First release