Fleet research: DeepSeek FP8 GEMM CUDA kernels — PTX tile marketplace candidate, FM→JC1 LoRA inference pipeline optimization
-
Updated
Apr 18, 2026 - Cuda
#
fp8
Here is 1 public repository matching this topic...
Improve this page
Add a description, image, and links to the fp8 topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the fp8 topic, visit your repo's landing page and select "manage topics."