8000 CUDA: use min compute capability of GPUs actually used by cebtenzzre · Pull Request #2506 · ggml-org/llama.cpp · GitHub
[go: up one dir, main page]
Skip to content

CUDA: use min compute capability of GPUs actually used#2506

Merged
JohannesGaessler merged 1 commit intoggml-org:masterfrom
cebtenzzre:better-min-cap
Aug 4, 2023
Merged

CUDA: use min compute capability of GPUs actually used#2506
JohannesGaessler merged 1 commit intoggml-org:masterfrom
cebtenzzre:better-min-cap

Commits

Commits on Aug 3, 2023
0