Files
llama.cpp/ggml
Sam/Samuel a31cf36ad9 metal : add opt_step_adamw and op_sum (#16529)
* scaffold to support opt step adamw on metal (not written so far)

* add opt-step-adamw kernel for metal

* pass op->src[4] as a separate buffer to the pipeline

* add bounds check to opt-step-adamw kernel

* complete scaffold for GGML_OP_SUM

* naive GGML_OP_SUM kernel

* remove unwanted comment

* change OP_SUM capability gate

* Add has_simdgroup_reduction to both ops to pass CI
2025-10-12 21:43:14 +03:00
..
2024-07-13 18:12:39 +02:00