]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
overview / pkg / ggml / sources / whisper.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 875d861)
vulkan: Use one row per workgroup for f32 mmv (llama/17711)
authorJeff Bolz <redacted>
Sat, 6 Dec 2025 10:12:26 +0000 (04:12 -0600)
committerGeorgi Gerganov <redacted>
Fri, 12 Dec 2025 15:53:20 +0000 (17:53 +0200)
commitc66c71e9f49180830715f6fe4d1b46c982bcb7c7
tree0b8504766d9874b3286d9fa8433f20828e89604ftree
parent875d8614733338c24d729de9f58df5d374f0f4dbcommit | diff
vulkan: Use one row per workgroup for f32 mmv (llama/17711)

The MoE models have a mul_mat_vec with very small m (32, 64, 128) right before
the topk_moe selection. Running multiple rows per wg doesn't utilize the SMs
well. I think even for larger m, f32 is so bandwidth-limited that running
multiple rows doesn't help.
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
Packaging of ggerganov/whisper.cpp