]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
overview / pkg / ggml / sources / whisper.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 3fee8a1)
opencl: route large q6_K lm_head to the flat GEMV (llama/26427)
authorHongqiang Wang <redacted>
Mon, 3 Aug 2026 14:36:19 +0000 (07:36 -0700)
committerGeorgi Gerganov <redacted>
Tue, 4 Aug 2026 10:37:47 +0000 (13:37 +0300)
commit4673c4bc32f2f65b8671e136d25d61337f48f10c
tree253b14a068350bd1488fbb944bc45910a2a4c350tree
parent3fee8a1e05a2ce904e12c6caf0e4c5b70701a6d2commit | diff
opencl: route large q6_K lm_head to the flat GEMV (llama/26427)

* add a direct size condition for `large` weights; the original
  dimension condition is insufficient -- q6_K lm_head for gemma-4 E2B
  has [1536, 262144], which is big enough to slowdown gemv_noshuffle but
  does not satisfy the dimension condition (ne0 >= 2048)
ggml/src/ggml-opencl/ggml-opencl.cpp diff | blob | history
Packaging of ggerganov/whisper.cpp