koboldcpp/scripts/snapdragon/adb
Max Krasnyansky c9872a2575
hexagon: HMX quantized matmul rework (#23368)
* hmx-mm: update debug logging in hmx-mm

* hmx-mm: update dequant logic to use HVX_vector_x2/4

* hmx-mm: remove non-pipelined version of the quantize matmul

It seems that we don't reall need non-pipelined version

* hmx-mm: use activation depth mode and update naming

Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>

* hex-mm: minor hmx matmul naming updates

* hmx-mm: remove unused vars

* snapdragon: scripts bump default ubatch-size to 1K

* hexagon: combine HMX and power and clock settings into a single set_power call

* hmx-mm: remove leftover of the scale repl helper

* hexagon: fix editconf error

---------

Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
2026-05-20 07:39:01 -07:00
..
llama-cli.farf Add experimental ggml-hexagon backend for the Hexagon NPU (#16547) 2025-10-22 13:47:09 -07:00
run-bench.sh hexagon: HMX quantized matmul rework (#23368) 2026-05-20 07:39:01 -07:00
run-cli.sh hexagon: HMX quantized matmul rework (#23368) 2026-05-20 07:39:01 -07:00
run-completion.sh hexagon: HMX quantized matmul rework (#23368) 2026-05-20 07:39:01 -07:00
run-mtmd.sh hexagon: HMX quantized matmul rework (#23368) 2026-05-20 07:39:01 -07:00
run-tool.sh hexagon: add support for basic and extended Op profiling (#22269) 2026-04-23 14:17:21 -07:00