]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 1d1d9a9)
Vulkan: route large matmuls to medium tile on Adreno (#24877)
authorRaman Shinde <redacted>
Sat, 11 Jul 2026 08:28:29 +0000 (13:58 +0530)
committerGitHub <redacted>
Sat, 11 Jul 2026 08:28:29 +0000 (10:28 +0200)
commit76f2798059575a96a12e4d34342165a4b6a6a312
tree3669df77778e7fa37f83fa7bd66bf4f29731e1d9tree
parent1d1d9a9ed7a4f09c4225ea4cc8fd3bd1cf2c940fcommit | diff
Vulkan: route large matmuls to medium tile on Adreno (#24877)

* [Vulkan] Fixes llama-cli breaking over longer promts sizes

The llama-cli was breaking for longer promts sizes for q4_0 quantized networks. Causing due to insufficient shared memory.

* Removed the un-used Adreno device

* Updated matmul for small pipeline.
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
Packaging of ggml-org/llama.cpp