]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
overview / pkg / ggml / sources / ggml / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 34e1e6c)
vulkan: handle quantize_q8_1 overflowing the max workgroup count (llama/18515)
authorJeff Bolz <redacted>
Mon, 5 Jan 2026 10:30:14 +0000 (04:30 -0600)
committerGeorgi Gerganov <redacted>
Sun, 11 Jan 2026 09:02:08 +0000 (11:02 +0200)
commit0624ce207f2436802441d7c56655b0496c5c139f
tree9b7744c16a41a88eb81fe1226e31e9fc16236e28tree
parent34e1e6cd3ad6ccfee4aeab735cc8540d5aa0b133commit | diff
vulkan: handle quantize_q8_1 overflowing the max workgroup count (llama/18515)

* vulkan: handle quantize_q8_1 overflowing the max workgroup count

* vulkan: Fix small tile size matmul on lavapipe

* fix mul_mat_id failures
src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
src/ggml-vulkan/vulkan-shaders/quantize_q8_1.comp diff | blob | history
tests/test-backend-ops.cpp diff | blob | history
Packaging of ggml-org/ggml