koboldcpp/docs
Max Krasnyansky b797b9ca81
hexagon: tensor repacking rework and support for tiled and flat quantized and fp weights (#30635)
* hex-repack: Remove need_repack hack and streamline the rules -- breaks GET_ROWS tests for K-Quants, will fix in the next round

Add support for GET_ROWS with tiled Q5_K
Add support for raw k-quant tensors

* hex-get-rows: vectorize raw k-quant ukernels

* hex-binary: refactor dma-loop thread functions (no need for macros, etc)

* hex-quant: move/cleanup flat and tiled dequant ukernels

* hex-repack: add support for repacking/tiling F16/F32/BF16 2D weights

* hex-get-rows: remove redundant cast

* hex-repack: proper filtering for BF16 that were leaking through during the tests

* hex-repack: streamling tiling rules (broken MM for now)

* hex-repack: fix mmid after tiling updates

* hex-repack: rename repacked to tiled in all flags/funcs/kernels/vars

* hex-mm: fix valid_rows calc in various kernels

* hex-mm: remove unused hvx_mm_nx_2d_flat

* hex-mm: cleanup and optimize hvx-tiled vec_dot_f16

* hex-mm: another pass at hvx-tiled f16 matmul

* hex-get-rows: optimize small src1

* hex-get-rows: one more pass at recovering get-rows perf regression

* hex-repack: export GGML_HEXAGON_RT_REPACK via run.py

* hex-mm-tiled: fix accuracy regression with hmx-tiled-f16 matmul

* hex-mm: fold tiled_f16 checks into the general flow

* hex-get-rows: remove dead checks for buffer_is_hexagon and tiling during supports_op

* hex-quant: proper tail handling in new dequant ukernels

* hex-get-rows: remove unused code and cleanup support checks

* hex-mm: cleanup ne1 padding/tiling checks

* hex-mm: cleanup fusion checks for tiled tensors

* hex-repack: add a little can_repack helper to reduce duplication

* hex-repack: never repack host buffers

* hex-repack: factor out repack functions into separate .h and .cpp

* hex-tensor: cleanup tensor ifaces for tiled/padded sizes

* hex-repack: update k-quant tiling comments and cleanup #defines

* hex-dcache: factor out l2/dcache dirty tracking and flushing

* hex-tensor: update docs and remove #defs duplication

* hex-tensor: remove unused/incorrect get_data interface

* hex-get-rows: cleanup vtcm layout build

* tests: disable huge MM tests

* hex-dcache: add missing header
2026-10-15 20:30:11 -07:00
..
android android: fix missing screenshots for Android.md (#18156) 2025-12-19 09:32:04 +02:00
backend hexagon: tensor repacking rework and support for tiled and flat quantized and fp weights (#30635) 2026-10-15 20:30:11 -07:00
development chat : name tool and argument parser rules by index (#30088) 2026-10-07 22:37:51 +02:00
multimodal model : support MiniCPM-V 4.7 (#29416) 2026-10-10 20:41:59 +02:00
ops docs : update zDNN supported ops (#30472) 2026-10-13 20:20:41 +02:00
android.md kleidiai : Update KleidiAI Documentation (#26078) 2026-09-01 10:45:13 +02:00
autoparser.md common : add LLM-jp-4.1 Harmony dialect handler (#29681) 2026-10-01 08:33:55 +03:00
build-profiling.md cmake : add PCH and unity build to improve build times (#28091) 2026-09-11 13:01:29 +02:00
build-riscv64-spacemit.md ggml-cpu : add Q8_0 IME1 matrix kernel for SpacemiT X60 (#28479) 2026-10-05 10:34:15 +03:00
build-s390x.md ggml-cpu(s390x): add Q1_0 vector intrinsic support (#28606) 2026-09-10 09:50:28 +03:00
build.md musa: drop mp_21 from the default architectures (#30203) 2026-10-09 12:51:24 +02:00
completions.md readme : refresh (#26280) 2026-07-30 16:14:37 +03:00
docker.md musa: build the docker images from the PH1 MUSA SDK image (#29481) 2026-09-26 21:50:27 +02:00
function-calling.md server: use port 9931 by default (#30159) 2026-10-09 10:40:03 +02:00
install.md docs: Adapt conda-forge package name (#26229) 2026-07-28 16:51:20 +02:00
llguidance.md llguidance build fixes for Windows (#11664) 2025-02-14 12:46:08 -08:00
models.md readme : refresh (#26280) 2026-07-30 16:14:37 +03:00
multi-gpu.md Write a readme on Multi-GPU usage in llama.cpp (#22729) 2026-05-07 17:48:40 +02:00
multimodal.md mtmd: refactor preprocessor, add mtmd_image_preproc_out (#24736) 2026-06-18 12:04:39 +02:00
ops.md docs : update zDNN supported ops (#30472) 2026-10-13 20:20:41 +02:00
preset.md common: add system-level config file (#26118) 2026-08-13 00:02:27 +02:00
release.md ci : allow make-release to target a specific commit (#27234) 2026-08-17 11:52:46 +03:00
speculative.md common : add --spec-draft-n-ubatch (#30459) 2026-10-14 13:58:05 +03:00
xcframework.md readme : refresh (#26280) 2026-07-30 16:14:37 +03:00