PR #15813 CUDA: Conv2d Tensor Core

CUDA: Conv2d Tensor Core #15813

mnehete32 wants to merge 3 commits into ggml-org:master from mnehete32:conv2d_tensor_core

CUDA: cov2d with tensor core

19596b17

CUDA: conv2d added comment

96db6275

github-actions added Nvidia GPU

github-actions added ggml

CUDA: conv2d support fp16 without wmma

2cd9fb0f

mnehete32 force pushed from 57aa09e0 to 2cd9fb0f 5 days ago

JohannesGaessler commented on 2025-09-05

Reviewers

JohannesGaessler

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone