SuperUserNameMan
SuperUserNameMan
Given that the generated text is random, what is the method to compare the output quality of q4_0 vs q4_k vs q6_k vs q8_0 in an objective (non-subjective) manner ?
@chigkim : thanks for these data. Here is my colorful interpretation of them :
> Mixtral-8x7B-Instruct-v0.1 - 42 sec (with heavy swapping) `Mixtral-8x7B-instruct-v0.1 Q4_K_M` is loaded in 14 secs on my Ryzen 5 5500u laptop (6 cores, 64GB RAM, Linux, SSD).
on my Linux it goes strait to 33 GB of ram :  **edit** : in case the version matters, i think the Mixtral instruct was from here : https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF...
Regarding GLFW's `glfwGetWindowContentScale()` : https://www.glfw.org/docs/latest/window_guide.html#window_scale > The content scale can be thought of as the ratio between the current DPI and the platform's default DPI. It is intended to be...
I think GLFW backend is correct.
The GLFW backend returns DPI scales factors. But the RGFW backend returns scaled resolutions. Shouldn't `GetWindowScaleDPI()` returns two scale factors instead of the scaled resolution of `platform.window->r` ? If yes,...
I tested both backends. With a 1280x720 window, here are the result of both on my Linux system : | Backend | `GetWindowScaleDPI()` | |-------------|-----------| | GLFW | `{ 1.145833...
just tested. It return `{ 1.0 , 1.0 }`. I don't think it has to return the exact same value as GLFW, because the GLFW documentation remains quite vague regarding...
No don't worry. I'm on Linux Mint with MATE desktop. I must have a default DPI hardcoded somewhere on my distro that explains my strange `{ 1.145833 , 1.145833 }`...