git.djapps.eu Git - pkg/ggml/sources/ggml/shortlog

]> git.djapps.eu Git - pkg/ggml/sources/ggml/shortlog

overview / pkg / ggml / sources / ggml / shortlog

2024-12-10	Jeff Bolz	vulkan: Add VK_NV_cooperative_matrix2 support for mul_m...	commit \| commitdiff \| tree
2024-12-08	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2024-12-08	Georgi Gerganov	scripts : update whisper sync	commit \| commitdiff \| tree
2024-12-05	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-12-05	Diego Devesa	ggml : add predefined list of CPU backend variants...	commit \| commitdiff \| tree
2024-12-05	Diego Devesa	ggml-cpu : fix HWCAP2_I8MM value (llama/10646)	commit \| commitdiff \| tree
2024-12-05	Jeff Bolz	vulkan: Implement "fast divide" (mul+shift) for unary...	commit \| commitdiff \| tree
2024-12-05	Nicolò Scipione	SYCL : Move to compile time oneMKL interface backend...	commit \| commitdiff \| tree
2024-12-05	Frankie Robertson	Avoid using __fp16 on ARM with old nvcc (llama/10616)	commit \| commitdiff \| tree
2024-12-05	Jeff Bolz	vulkan: optimize and reenable split_k (llama/10637)	commit \| commitdiff \| tree
2024-12-04	PAB	ggml: add `GGML_SET` Metal kernel + i32 CPU kernel...	commit \| commitdiff \| tree
2024-12-03	PAB	ggml : add `GGML_PAD_REFLECT_1D` operation (#1034)	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ci : update requirements.txt	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ci : remove opencl workflow	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ci : fix pip env	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	files : remove make artifacts	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	common : fix compile warning	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	authors : update	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-12-03	Diego Devesa	ggml : move AMX to the CPU backend (llama/10570)	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	scripts : remove amx from sync	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	metal : small-batch mat-mul kernels (llama/10581)	commit \| commitdiff \| tree
2024-12-03	Akarshan Biswas	SYCL: Fix and switch to GGML_LOG system instead of...	commit \| commitdiff \| tree
2024-12-03	Adrien Gallouët	ggml-cpu: replace AArch64 NEON assembly with intrinsics...	commit \| commitdiff \| tree
2024-12-03	Eve	vulkan: Dynamic subgroup size support for Q6_K mat_vec...	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ggml : fix I8MM Q4_1 scaling factor conversion (llama...	commit \| commitdiff \| tree
2024-12-03	Shupei Fan	ggml-cpu: fix typo in gemv/gemm iq4_nl_4_4 (llama/10580)	commit \| commitdiff \| tree
2024-12-03	Alberto Cabrera...	sycl : offload of get_rows set to 0 (llama/10432)	commit \| commitdiff \| tree
2024-12-03	Alberto Cabrera...	sycl : Reroute permuted mul_mats through oneMKL (llama...	commit \| commitdiff \| tree
2024-12-03	Chenguang Li	CANN: RoPE operator optimization (llama/10563)	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: get the first command buffer submitted sooner...	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ggml : remove redundant copyright notice + update authors	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	ggml : fix row condition for i8mm kernels (llama/10561)	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	cmake : fix ARM feature detection (llama/10543)	commit \| commitdiff \| tree
2024-12-03	Shupei Fan	ggml-cpu: support IQ4_NL_4_4 by runtime repack (llama...	commit \| commitdiff \| tree
2024-12-03	Sergio López	kompute : improve backend to pass test_backend_ops...	commit \| commitdiff \| tree
2024-12-03	leo-pony	CANN: Fix SOC_TYPE compile bug (llama/10519)	commit \| commitdiff \| tree
2024-12-03	Chenguang Li	CANN: ROPE operator optimization (llama/10540)	commit \| commitdiff \| tree
2024-12-03	uvos	Add some minimal optimizations for CDNA (llama/10498)	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	metal : fix group_norm support condition (llama/0)	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: define all quant data structures in types.comp...	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: Handle GPUs with less shared memory (llama...	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: further optimize q5_k mul_mat_vec (llama/10479)	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: skip integer div/mod in get_offsets for batch_i...	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: optimize Q2_K and Q3_K mul_mat_vec (llama/10459)	commit \| commitdiff \| tree
2024-12-03	R0CKSTAR	mtgpu: Add MUSA_DOCKER_ARCH in Dockerfiles && update...	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: fix group_norm (llama/10496)	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	cmake : enable warnings in llama (llama/10474)	commit \| commitdiff \| tree
2024-12-03	Charles Xu	ggml-cpu: cmake add arm64 cpu feature check for macos...	commit \| commitdiff \| tree
2024-12-03	Shanshan Shen	CANN: Improve the Inferencing Performance for Ascend...	commit \| commitdiff \| tree
2024-12-03	Chenguang Li	CANN: RoPE and CANCAT operator optimization (llama...	commit \| commitdiff \| tree
2024-12-03	Junil Kim	vulkan: Fix a vulkan-shaders-gen arugment parsing error...	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	metal : enable mat-vec kernels for bs <= 4 (llama/10491)	commit \| commitdiff \| tree
2024-12-03	Diego Devesa	llama : accept a list of devices to use to offload...	commit \| commitdiff \| tree
2024-12-03	Diego Devesa	ggml : add support for dynamic loading of backends...	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	tests : fix compile warning	commit \| commitdiff \| tree
2024-12-03	Georgi Gerganov	metal : minor code formatting	commit \| commitdiff \| tree
2024-12-03	Diego Devesa	ggml : do not use ARM features not included in the...	commit \| commitdiff \| tree
2024-12-03	leo-pony	CANN: Support Ascend310P to accelerate F32 and F16...	commit \| commitdiff \| tree
2024-12-03	Diego Devesa	cuda : optimize argmax (llama/10441)	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: predicate max operation in soft_max shaders...	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: copy iq4_nl LUT into shared memory (llama/10409)	commit \| commitdiff \| tree
2024-12-03	Jeff Bolz	vulkan: further optimize mul_mat_vec using larger loads...	commit \| commitdiff \| tree
2024-12-03	haopeng	add cmake rvv support (llama/10411)	commit \| commitdiff \| tree
2024-12-03	mahorozte	CUDA: remove unnecessary warp reduce in FA (#1032)	commit \| commitdiff \| tree
2024-12-02	PAB	feat: add `GGML_UNARY_OP_ARGMAX` Metal kernel (#1019)	commit \| commitdiff \| tree
2024-11-28	PAB	metal : add `GGML_OP_CONV_TRANSPOSE_1D` kernels (#1026)	commit \| commitdiff \| tree
2024-11-27	Johannes Gäßler	examples: link to HuggingFace mirror of MNIST data...	commit \| commitdiff \| tree
2024-11-26	Tristan Druyen	Fix build docs for hip (#1029)	commit \| commitdiff \| tree
2024-11-26	Frankie Robertson	Do not include arm_neon.h when compiling CUDA code...	commit \| commitdiff \| tree
2024-11-20	M Refi D.A	Create .gitmodules for the kompute backend (#1024)	commit \| commitdiff \| tree
2024-11-20	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2024-11-20	slaren	ggml/sched : do not skip views in pre-assignments	commit \| commitdiff \| tree
2024-11-20	Johannes Gäßler	ggml-opt: fix data corruption (#1022)	commit \| commitdiff \| tree
2024-11-19	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-19	bandoti	Add required ggml-base and backend libs to cmake pkg...	commit \| commitdiff \| tree
2024-11-19	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-19	Diego Devesa	cuda : fix CUDA_FLAGS not being applied (llama/10403)	commit \| commitdiff \| tree
2024-11-19	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-19	Romain Biessy	sycl : Add option to set the SYCL architecture for...	commit \| commitdiff \| tree
2024-11-19	Jeff Bolz	vulkan: Optimize soft_max (llama/10301)	commit \| commitdiff \| tree
2024-11-19	Alberto Cabrera...	sycl: Revert MUL_MAT_OP support changes (llama/10385)	commit \| commitdiff \| tree
2024-11-19	Diego Devesa	cuda : only use native when supported by cmake (llama...	commit \| commitdiff \| tree
2024-11-19	Jeff Bolz	vulkan: remove use of null initializer (llama/10372)	commit \| commitdiff \| tree
2024-11-18	Plamen Minev	metal : fox offset integer overflows in im2col (#1015)	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-18	0cc4m	Vulkan: Fix device info output format specifiers (llama...	commit \| commitdiff \| tree
2024-11-18	PAB	metal : add `GGML_UNARY_OP_ELU` kernel (#1018)	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-18	Johannes Gäßler	CUDA: fix MMV kernel being used for FP16 src1 (llama...	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2024-11-18	Johannes Gäßler	CMake: fix typo in comment [no ci] (llama/10360)	commit \| commitdiff \| tree
2024-11-18	Diego Devesa	llama : only use default buffer types for the KV cache...	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	metal : refactor kernel args into structs (llama/10238)	commit \| commitdiff \| tree
2024-11-18	FirstTimeEZ	ggml : fix undefined reference to 'getcpu' (llama/10354)	commit \| commitdiff \| tree
2024-11-18	Johannes Gäßler	CUDA: remove DMMV, consolidate F16 mult mat vec (llama...	commit \| commitdiff \| tree
2024-11-18	Johannes Gäßler	CMake: default to -arch=native for CUDA build (llama...	commit \| commitdiff \| tree
2024-11-18	Diego Devesa	ggml : fix possible buffer use after free in sched...	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	ggml : inttypes.h -> cinttypes (llama/0)	commit \| commitdiff \| tree
2024-11-18	Georgi Gerganov	ggml : adapt AMX to tensor->grad removal (llama/0)	commit \| commitdiff \| tree
next

Packaging of ggml-org/ggml

RSS Atom