nicolas
/
ik_llama_opt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
ik_llama_opt/ggml/src
History
Kawrakow bd342d624f
DS4 optimizations (#2169) 
* Adding ds4_comp op with CPU implementation

* ds4_comp on CUDA

* ds4_comp: ratio = 4 specialization

Surprisingly small performance gain

* Also handle HCA via ds4_comp

But much smaller gain, if any.

* Delete commented out stuff

* Remove the [(size_t) il] noise

* Minor

* Fix quantized cache
 		2026-07-25 08:52:38 +03:00
..
cmake Merge vulkan code from mainline up to commit of 6/28/2025 (#563) 2025-07-02 08:49:42 +02:00
ggml-cann Merge mainline - Aug 12 2024 (#17) 2024-08-12 15:14:32 +02:00
ggml-cuda DS4 optimizations (#2169) 2026-07-25 08:52:38 +03:00
ggml-sycl Merge mainline - Aug 12 2024 (#17) 2024-08-12 15:14:32 +02:00
iqk DS4: slowly approaching a meaningful performance (#2165) 2026-07-22 17:18:57 +03:00
llamafile Merge mainline llama.cpp (#3) 2024-07-27 07:55:01 +02:00
vulkan-shaders Port of Qwen3-VL support from mainline (#883) 2025-11-04 19:20:54 +02:00
CMakeLists.txt Fix build on macOS (#2137) 2026-07-17 14:05:55 +03:00
ggml-aarch64.c Merge mainline - Aug 12 2024 (#17) 2024-08-12 15:14:32 +02:00
ggml-aarch64.h Merge mainline llama.cpp (#3) 2024-07-27 07:55:01 +02:00
ggml-alloc.c Support for Qwen 3.5 MTP (dense models only) (#1698) 2026-04-28 07:47:50 +02:00
ggml-backend-impl.h Merge vulkan code from mainline up to commit of 6/28/2025 (#563) 2025-07-02 08:49:42 +02:00
ggml-backend.cpp openpangu: fused latent attention op (GGML_OP_LATENT_ATTN) (#2168) 2026-07-23 14:50:45 +03:00
ggml-cann.cpp Merge vulkan code from mainline up to commit of 6/28/2025 (#563) 2025-07-02 08:49:42 +02:00
ggml-common.h Bonsai support (AVX2, generic) (#1570) 2026-04-02 16:54:08 +02:00
ggml-cuda.cu DS4 optimizations (#2169) 2026-07-25 08:52:38 +03:00
ggml-impl.h MXFP4 (#682) 2025-08-09 08:40:18 +03:00
ggml-metal.m metal: implement ROPE_MULTI (mrope/imrope) kernels (#2140) 2026-07-17 17:49:13 +03:00
ggml-metal.metal metal: implement ROPE_MULTI (mrope/imrope) kernels (#2140) 2026-07-17 17:49:13 +03:00
ggml-moe-prefetch.cpp Add --prefetch-experts to stream mmap'd MoE experts into page cache (#2101) 2026-07-11 10:43:42 +03:00
ggml-moe-prefetch.h Add --prefetch-experts to stream mmap'd MoE experts into page cache (#2101) 2026-07-11 10:43:42 +03:00
ggml-quants.c Use AVX version VNNI intrinsic when AVX512VNNI not available. (#1748) 2026-05-09 09:02:06 +03:00
ggml-quants.h Quantization options (#1677) 2026-04-23 09:05:39 +02:00
ggml-rpc.cpp server: improve speed of speculative decoding (#1119) 2026-01-10 08:01:22 +02:00
ggml-sycl.cpp Merge vulkan code from mainline up to commit of 6/28/2025 (#563) 2025-07-02 08:49:42 +02:00
ggml-vulkan.cpp Port of Qwen3-VL support from mainline (#883) 2025-11-04 19:20:54 +02:00
ggml.c DS4 optimizations (#2169) 2026-07-25 08:52:38 +03:00