forked from ggml-org/llama.cpp
-
Notifications
You must be signed in to change notification settings - Fork 38
Pull requests: tetherto/qvac-fabric-llm.cpp
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add fused DeepSeek4 optimization ops
Apple Metal
CUDA
ggml
model
testing
Vulkan
#186
opened Jul 21, 2026 by
gianni-cor
Loading…
[testing] standalone qwen3vl encoder + CPU repack (do not merge)
documentation
Improvements or additions to documentation
mtmd
llama-model : load QKV bias into wq_b/wk_b/wv_b/wqkv_b (not dead bq/b…
#173
opened Jun 26, 2026 by
zoq
Loading…
opencl : fix Q8 tensor readback from generic buffers
ggml
OpenCL
#164
opened Jun 19, 2026 by
gianni-cor
•
Draft
perf: ARM CPU conv kernels for DocTR detection (rebased onto temp-9341)
ggml
testing
#162
opened Jun 19, 2026 by
olyasir
Loading…
Enable Android Vulkan int dot builds
android
Apple Metal
build
CUDA
devops
documentation
Improvements or additions to documentation
examples
ggml
python
script
server
testing
Vulkan
#151
opened Jun 14, 2026 by
watsoncsulahack
Loading…
ci: fix Vulkan Docker build failures and GHCR registry timeout
build
devops
#136
opened May 18, 2026 by
Ektisad25
Loading…
cuda: restrict out_prod support to f32 inputs
ggml
Nvidia GPU
#110
opened Mar 19, 2026 by
GuthL
Loading…
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.