rtx-4050

Star

Here are 2 public repositories matching this topic...

Sharveswar007 / SSBLAST

Star

First open-source FP8 linear solver for consumer NVIDIA GPUs — 2-3x faster than cuBLAS FP64. pip install ssblast

python machine-learning hpc gpu cuda nvidia triton cupy numerical-computing fp8 linear-solver rtx-4050

Updated Jun 5, 2026
Python

ChinmayHudedamani / edge-llm-runtime

Star

Chinmay's AI Assistant (edge-llm-runtime) is a 100% offline, high-efficiency local inference engine for NVIDIA consumer GPUs featuring group-wise INT4 weight quantization (3.39x memory reduction), register-level fused dequantization GEMM kernels, interactive persona modes, and a publication-grade PDF/charting tools suite.

cuda pytorch inference-engine edge-ai llm paged-attention int4-quantization rtx-4050

Updated Sep 14, 2026
HTML

Add this topic to your repo

To associate your repository with the rtx-4050 topic, visit your repo's landing page and select "manage topics."

Learn more