* maybe it worth showing some feedback when AI is thinking/composing the answer vs. when it's already finished. Currently, on my machine words appear rather slowly, so it's not always clear if Assistant is done answering or not. Currently, I use CPU fan noise to check when it's done thinking :)
* also, scrolling the white text field to the bottom might be useful to see that text appeared.
* and it looks like it has some issues with non-latin symbols - try submitting some emoji or cyrillic text to it - black console at the bottom shows "Invalid UTF-8 leading byte 0x0000008d encountered when deserializing a UTF-8 string in wasm memory to a JS string!" message and user input (and AI output) appears garbled in the white field at the top.
A bit of feedback, if I may:
* maybe it worth showing some feedback when AI is thinking/composing the answer vs. when it's already finished. Currently, on my machine words appear rather slowly, so it's not always clear if Assistant is done answering or not. Currently, I use CPU fan noise to check when it's done thinking :)
* also, scrolling the white text field to the bottom might be useful to see that text appeared.
* and it looks like it has some issues with non-latin symbols - try submitting some emoji or cyrillic text to it - black console at the bottom shows "Invalid UTF-8 leading byte 0x0000008d encountered when deserializing a UTF-8 string in wasm memory to a JS string!" message and user input (and AI output) appears garbled in the white field at the top.
Other than that, great stuff!