zakkor

icon location X should be designed not by piling feature on top of feature, but by removing the weaknesses and restrictions that make additional features appear necessary.

Results 15 comments of zakkor

How to connect Lluminous to local llama.cpp?

@vojtapolasek there is now Ollama support, just start the Ollama server and you should see your installed models in the model picker.

ggml : unified CMake build

Not sure if this is the right place to put it, but: Before, I could clone the project, run `make`, and everything worked perfectly -- it also automatically chose the...

Feature Request: `reasoning_effort` parameter for reasoning models like R1

Indeed, system prompts would be a way of doing this, although I don't believe they can guarantee that you'll ALWAYS get the model to generate at least a minimum amount...

Feature Request: `reasoning_effort` parameter for reasoning models like R1

Bump

Can you improve the visual design of the extensions?

This is in Jetbrains IntelliJ. Here's a couple of UI/UX issues I found within a few minutes of using it (in no particular order): 1. When opening the Continue panel...