llama.cpp
vulkan: Allow fallback to sysmem memory when vidmem is full
#15649

Merged

vulkan: Allow fallback to sysmem memory when vidmem is full #15649

0cc4m merged 2 commits into ggml-org:master from jeffbolznv:sysmem_fallback

vulkan: Allow fallback to sysmem memory when vidmem is full

992383cc

jeffbolznv requested a review from

0cc4m 15 days ago

github-actions added Vulkan

github-actions added ggml

vulkan: Add env var GGML_VK_ALLOW_SYSMEM_FALLBACK

be2d914a

0cc4m approved these changes on 2025-08-31

0cc4m merged b97c9edc into master 13 days ago

0cc4m commented on 2025-09-09

Reviewers

0cc4m

Assignees

No one assigned

Labels

Vulkan ggml

Milestone

No milestone