Fast LLM speculative inference server for consumer hardware.
kernel cuda cuda-kernels nvidia-cuda luce rtx3090 llama-cpp local-ai qwen speculative-decoding dflash megakernel speculative-prefill pflash lucebox
-
Updated
Jun 13, 2026 - C++
#
luce
Here are 2 public repositories matching this topic...
Language: All
Filter by language
Improve this page
Add a description, image, and links to the luce topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the luce topic, visit your repo's landing page and select "manage topics."