]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: a678916)
ggml-webgpu: updated matrix-vector multiplication (#21738)
authorneha-ha <redacted>
Mon, 20 Apr 2026 14:37:17 +0000 (07:37 -0700)
committerGitHub <redacted>
Mon, 20 Apr 2026 14:37:17 +0000 (07:37 -0700)
commita6cc43c286a2ebc429aa69b9a4d16de082cedb51
tree46c70865911d7c1e70030b9934e7faa410a3898btree
parenta678916623ddef89c2a43776df24e00a52b17638commit | diff
ggml-webgpu: updated matrix-vector multiplication (#21738)

* merged properly, but slow q3_k and q5_k with u32 indexing

* Start on new mat-vec

* New format float paths working

* Working q4_0

* Work on remaining legacy q-types

* port k-quants to new matvec

* remove old shader

* Remove old constants, format

* remove accidental file

---------

Co-authored-by: Neha Abbas <redacted>
Co-authored-by: Reese Levine <redacted>
ggml/src/ggml-webgpu/ggml-webgpu-shader-lib.hpp diff | blob | history
ggml/src/ggml-webgpu/ggml-webgpu.cpp diff | blob | history
ggml/src/ggml-webgpu/wgsl-shaders/common_decls.tmpl diff | blob | history
ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_vec.wgsl diff | blob | history
Packaging of ggml-org/llama.cpp