Louis

cc @tikikun ``` [NITRO]::Error: .......GGML_ASSERT: /nitro/llama.cpp/ggml-backend.c:1270: (char *)addr + ggml_backend_buffer_get_alloc_size(buffer, tensor)

> Sure, like I said I already ran nightly - here's the logfile it generated: > > https://gist.github.com/iwootten/6cce37433abaaad3f3df039c50a496e2 > > Looks like it's failing to load the model. This seems...

I'm sorry to ask, but could you please try again with the latest release? We have recently fixed some issues and updated the core engine.

Token Speed = Tokens / second (time window), there is no streaming gap so the Token Speed does not mean much in this case. The response will be returned at...

Since this is a misleading feature request, not really a bug. We can close it for now.

Thank you for the great contribution! @markmehere.

This is our recent update to avoid some misleading issues, such as: - Users can input a very large value, which breaks the app. - Users can set the maximum...

> Yeah this is kinda weird, i cannot clear the number @urmauur > > Screen.Recording.2024-01-14.at.10.54.51.PM.mov @imtuyethan that's is minimum threshold

Experimental feature: 0.4.7

Thank you for the first check @urmauur. Now is my turn for the nitro crash & error handling.