llama.cpp
CUDA: Allow concurrent streams per split for multi-GPU
#28198

Merged

CUDA: Allow concurrent streams per split for multi-GPU #28198

am17an merged 2 commits into ggml-org:master from tannerbruhn:cuda-graph-opt-multi-gpu

CUDA: Allow CUDA optimization per split for multi-GPU.

044a1d6d

tannerbruhn requested a review 43 days ago

github-actions added ggml

github-actions added CUDA

am17an commented on 2026-09-02

am17an approved these changes on 2026-09-02

Update ggml/src/ggml-cuda/ggml-cuda.cu

3f4500f9

am17an added merge ready

CISC approved these changes on 2026-09-03

am17an merged 0ba6499c into master 42 days ago

tannerbruhn deleted the cuda-graph-opt-multi-gpu branch 41 days ago

Reviewers

CISC

am17an

Assignees

No one assigned

Labels

ggml merge ready CUDA

Milestone

No milestone