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

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

overview / pkg / ggml / sources / ggml / shortlog

2025-09-09	distlibs	gitignore : ignore idea files (#1339)	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-09-05	Gabe Goodhart	metal : Add template specialization for mul_mm_id w...	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: Refactor ND to NZ workspace to be per-device...	commit \| commitdiff \| tree
2025-09-05	leejet	ggml: add ops for WAN video model (cuda && cpu) (llama...	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Fix precision issue on 310I DUO multi-devices...	commit \| commitdiff \| tree
2025-09-05	rmatif	opencl: add hs=40 to FA (llama/15758)	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: fix acl_rstd allocation size in ggml_cann_rms_nor...	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: fix mmv subgroup16 selection (llama/15775)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: don't use std::string in load_shaders, to impro...	commit \| commitdiff \| tree
2025-09-05	Daniel Bevenius	vulkan : update ggml_vk_instance_validation_ext_availab...	commit \| commitdiff \| tree
2025-09-05	Shin-myoung...	ggml vulkan: add hardsigmoid and hardswish operations...	commit \| commitdiff \| tree
2025-09-05	Oliver Simons	CUDA: Optimize `rms_norm_f32` kernel and its fused...	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Add RoPE contiguous check for 310I DUP device...	commit \| commitdiff \| tree
2025-09-05	xctan	ggml-cpu : optimize RVV kernels (llama/15720)	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Mask unsupported TRANSPOSE_1D operator (llama...	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: Fix type float_t to float (llama/15736)	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: fix shaders gen when no integer dot is availabl...	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Resolve soft_max precision issue (llama/15730)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Fix macro parameter order for f32 matmul shader...	commit \| commitdiff \| tree
2025-09-05	rmatif	opencl: add attn sinks support for FA kernels (llama...	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: Support eager execution mode under ACL graph...	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Support ext_factor in rope (llama/15710)	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	ggml-backend: raise GGML_MAX_SPLIT_INPUTS (llama/15722)	commit \| commitdiff \| tree
2025-09-05	Gilad S.	vulkan: use memory budget extension to read memory...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: add missing clamps in new mul_mat_id paths...	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: disable large mmv subgroups on older Nvidia...	commit \| commitdiff \| tree
2025-09-05	s-goto-11	ggml: SVE support for exponential functions (llama...	commit \| commitdiff \| tree
2025-09-05	Prashant Vithule	ggml: aarch64: Implement SVE F16 kernels for vector...	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	Vulkan: Add Integer Dot Product mul_mat_vec shader...	commit \| commitdiff \| tree
2025-09-05	Daniel Bevenius	ggml : WebGPU add TRANSPOSE and RESHAPE to supported...	commit \| commitdiff \| tree
2025-09-05	Akarshan Biswas	CUDA: fix build error from ambiguous __half conversions...	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: Optimize MUL_MAT_ID (llama/15658)	commit \| commitdiff \| tree
2025-09-05	hipudding	CANN: fix RoPE cache issue on multi-device (llama/15629)	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	metal : fix checks for available FA kernels (llama...	commit \| commitdiff \| tree
2025-09-05	Diego Devesa	llama : separate compute buffer reserve from fattn...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: handle large sizes for get_rows (llama/15686)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: mul_mat_id coopmat2 optimizations (llama/15546)	commit \| commitdiff \| tree
2025-09-05	Daniel Bevenius	vulkan : remove unused portability_enumeration_ext...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Allow fallback to sysmem memory when vidmem...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: clamp matmul and FA results to the max finite...	commit \| commitdiff \| tree
2025-09-05	Charles Xu	ggml: update kleidiai to v1.13.0 (llama/15663)	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	llama: use FA + max. GPU layers by default (llama/15434)	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: use FP32 arithmetic for conv2d (llama/15683)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Skip syncing for prealloc_y when it is reused...	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: FIx compiler warnings (llama/15661)	commit \| commitdiff \| tree
2025-09-05	Aman Gupta	CUDA: fix bug in rms_norm fusion (llama/15660)	commit \| commitdiff \| tree
2025-09-05	Aman Gupta	CUDA: fuse adds, fuse add with rms norm (llama/15631)	commit \| commitdiff \| tree
2025-09-05	mnehete32	CUDA: add conv2d (llama/15635)	commit \| commitdiff \| tree
2025-09-05	Aaron Teo	ggml-cpu: fix invalid hsum build in debug s390x (llama...	commit \| commitdiff \| tree
2025-09-05	compilade	ggml : fix SSM_SCAN for n_groups > 1 (llama/15625)	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	kv-cache : remove LLAMA_SET_ROWS checks (llama/15505)	commit \| commitdiff \| tree
2025-09-05	matiaslin	cuda: Add cublasLt_static linking when GGML_STATIC...	commit \| commitdiff \| tree
2025-09-05	uvos	HIP: Enable support for ggml_backend_cuda_register_host...	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: refactor mask handling and improve performance...	commit \| commitdiff \| tree
2025-09-05	xctan	ggml-cpu : add basic RVV support for vector f32 ops...	commit \| commitdiff \| tree
2025-09-05	rmatif	OpenCL: add fused group_norm/norm, mul, add (llama...	commit \| commitdiff \| tree
2025-09-05	Diego Devesa	tests : fix test-opt with GGML_BACKEND_DL (llama/15599)	commit \| commitdiff \| tree
2025-09-05	Akarshan Biswas	SYCL: fix rms_norm_mul_add for tensor dim not a multipl...	commit \| commitdiff \| tree
2025-09-05	Eve	tests: add performance test for mul mat id (llama/15543)	commit \| commitdiff \| tree
2025-09-05	shalinib-ibm	llamafile: PowerPC Sgemm Optimization (llama/15558)	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: return -1 for nonexistent compiled arch (llama...	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	metal : optimize FA vec for large sequences and BS...	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	metal : improve `MUL_MAT_ID` (llama/15541)	commit \| commitdiff \| tree
2025-09-05	Sigbjørn Skjæret	metal : remove contiguous assertion for src0 in IM2COL...	commit \| commitdiff \| tree
2025-09-05	Yoshi_likes_e4	Add a warning for special devices (llama/15563)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Remove splitting for mul_mat_id (llama/15568)	commit \| commitdiff \| tree
2025-09-05	Qeeweew	CUDA: Accelerate MXFP4 table lookup using `__byte_perm...	commit \| commitdiff \| tree
2025-09-05	lhez	opencl: fix support ops condition for `rms_norm` (llama...	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: fix min subgroup 16 condition for mmid subgroup...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	tests: Generate unique input values for count_equal...	commit \| commitdiff \| tree
2025-09-05	Ihar Hrachyshka	metal: fix regression when no metal devices are present...	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: MoE helper in device code, better tile sizes...	commit \| commitdiff \| tree
2025-09-05	Georgi Gerganov	metal : add FA kernels for HS=40 (llama/15559)	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: ROPE cache sin/cos repeat (llama/15501)	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: apply MUL_MAT_ID subgroup optimization to non...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Support FA with any multiple of 8 head sizes...	commit \| commitdiff \| tree
2025-09-05	Ruben Ortlam	vulkan: enable Conv2D for Apple after MoltenVK fixed...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: workaround MoltenVK compile failure in multi_ad...	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: fix half2 -> half conversion for HIP (llama/15529)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: optimize rms_norm, and allow the work to spread...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Rewrite synchronization to allow some overlap...	commit \| commitdiff \| tree
2025-09-05	Acly	vulkan : support ggml_mean (llama/15393)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: optimize mul_mat_id loading row ids into shared...	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	test-opt: allow slight inprecision (llama/15503)	commit \| commitdiff \| tree
2025-09-05	Reese Levine	ggml WebGPU: add support for quantization types (llama...	commit \| commitdiff \| tree
2025-09-05	rmatif	ggml: add `conv3d` op (llama/15182)	commit \| commitdiff \| tree
2025-09-05	Yavor Ivanov	cuda : add Pad Reflect 1D support (llama/14659)	commit \| commitdiff \| tree
2025-09-05	Aaron Teo	ggml-cpu: Support Q5_0 and Q5_1 on s390x (llama/15486)	commit \| commitdiff \| tree
2025-09-05	Chenguang Li	CANN: Optimize RMS_NORM using cache (llama/15419)	commit \| commitdiff \| tree
2025-09-05	Diego Devesa	sched : fix possible use of wrong ids tensor when offlo...	commit \| commitdiff \| tree
2025-09-05	Acly	vulkan : support conv_2d_dw with f16 weights (llama...	commit \| commitdiff \| tree
2025-09-05	Dong Won Kim	vulkan: add exp operation (llama/15456)	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: Reuse conversion results in prealloc_y (llama...	commit \| commitdiff \| tree
2025-09-05	Xuan-Son Nguyen	ggml : fix condition of im2col on Metal backend (llama...	commit \| commitdiff \| tree
2025-09-05	R0CKSTAR	musa: add GGML_UNUSED_VARS (llama/15446)	commit \| commitdiff \| tree
2025-09-05	Diego Devesa	sched : copy only the used experts when offloading...	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: refactor FA support/selection code (llama/15454)	commit \| commitdiff \| tree
2025-09-05	Johannes Gäßler	CUDA: replace GGML_CUDA_F16 with CUDA arch checks ...	commit \| commitdiff \| tree
2025-09-05	Jeff Bolz	vulkan: shorten pipeline name strings (llama/15431)	commit \| commitdiff \| tree
next

Packaging of ggml-org/ggml

RSS Atom