dp4a

Here are 4 public repositories matching this topic...

qr0n / p100-llama-cpp

pascal inference p100 inference-optimization llama-cpp dp4a

Updated Sep 24, 2026
C++

jajmangold / superl8

DP4A FlashAttention-2 and GEMM for Volta GPUs. 46 TOP/s INT8 on CMP 100-210 where tensor cores are firmware-disabled. The .superl8 format loads weights at memory speed.

cuda inference pytorch quantization gemm int8 volta flash-attention sm70 dp4a

Updated Sep 22, 2026
Python

townsendmerino / wgpu

Star

Minimal compute-only Go (CGO) binding over wgpu-native v29 — a drop-in for the cogentcore/webgpu subset, with full control over the WGSL dot4I8Packed builtin.

go metal gpu cgo compute webgpu wgsl wgpu-native dp4a

Updated Jul 16, 2026
C++

A fused descriptor matcher for COLMAP: 1.28x faster SIFT matching with byte-identical output and 55% less matcher VRAM, plus an accounting of why that does not reach end-to-end runtime. Patch overlay, no vendored COLMAP source.

structure-from-motion cuda sift colmap gpu-optimization dp4a

Updated Aug 20, 2026
TeX

Add this topic to your repo

To associate your repository with the dp4a topic, visit your repo's landing page and select "manage topics."

Learn more

Navigation Menu