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

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

overview / pkg / ggml / sources / ggml / shortlog

2026-01-13	Jeff Bolz	vulkan: change memory_logger to be controlled by an...	commit \| commitdiff \| tree
2026-01-13	Jeff Bolz	vulkan: Use VK_EXT_shader_64bit_indexing to handle...	commit \| commitdiff \| tree
2026-01-13	Ruben Ortlam	vulkan: Disable large coopmat matmul configuration...	commit \| commitdiff \| tree
2026-01-13	Ruben Ortlam	Vulkan: Optimize Matmul parameters for AMD GPUs with...	commit \| commitdiff \| tree
2026-01-11	Georgi Gerganov	sync : llma.cpp	commit \| commitdiff \| tree
2026-01-11	shaofeiqi	opencl: add SOFTPLUS op support (llama/18726)	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	test-backend-ops: fix mxfp4 tests on blackwell (llama...	commit \| commitdiff \| tree
2026-01-11	Johannes Gäßler	HIP: adjust RDNA3.5 MMQ kernel selction logic (llama...	commit \| commitdiff \| tree
2026-01-11	Perry Naseck	cmake : update blas logic (llama/18205)	commit \| commitdiff \| tree
2026-01-11	Michael Wand	Corrected: changed s13 = src1->nb[3] instead of nb...	commit \| commitdiff \| tree
2026-01-11	shaofeiqi	opencl: add EXPM1 op (llama/18704)	commit \| commitdiff \| tree
2026-01-11	Reese Levine	Updates to webgpu get_memory (llama/18707)	commit \| commitdiff \| tree
2026-01-11	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2026-01-11	Aaron Teo	llama: use host memory if device reports 0 memory ...	commit \| commitdiff \| tree
2026-01-11	Masashi Yoshimura	ggml-webgpu: Fix GGML_MEM_ALIGN to 8 for emscripten...	commit \| commitdiff \| tree
2026-01-11	Reese Levine	ggml webgpu: initial flashattention implementation...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: fix push constant size for quantize_q8_1 (llama...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: optimize ssm_scan (llama/18630)	commit \| commitdiff \| tree
2026-01-11	도로로도로또	metal : add MoE kernel specialization for ne20=5 (llama...	commit \| commitdiff \| tree
2026-01-11	Doctor Shotgun	ggml: add env var GGML_OP_OFFLOAD_MIN_BATCH (llama...	commit \| commitdiff \| tree
2026-01-11	shaofeiqi	opencl: add FILL op support (llama/18682)	commit \| commitdiff \| tree
2026-01-11	Oliver Walsh	cuda : fix build on cuda 12.8 (llama/18672)	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: reject ops when a tensor is too large to alloca...	commit \| commitdiff \| tree
2026-01-11	virajwad	vulkan: Warptile tuning for Intel Xe2/Xe3 (llama/18178)	commit \| commitdiff \| tree
2026-01-11	Eve	vulkan: more mul mat optimizations (llama/18533)	commit \| commitdiff \| tree
2026-01-11	hipudding	CANN: Fix rename for get_env (llama/18652)	commit \| commitdiff \| tree
2026-01-11	Raul Torres	CANN: Rename `get_env` to `get_env_as_lowercase` (llama...	commit \| commitdiff \| tree
2026-01-11	Max Krasnyansky	Hexagon add support for f16/f32 flash attention, scale...	commit \| commitdiff \| tree
2026-01-11	Aadeshveer...	ggml : optimize cuda ssm_scan using warp-level reductio...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: support buffer_from_host_ptr (llama/18467)	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	ggml-cuda: refactor cuda graph usage (llama/18637)	commit \| commitdiff \| tree
2026-01-11	Beinsezii	mmq.cu: tune mmq/rocblas switching for RDNA (llama...	commit \| commitdiff \| tree
2026-01-11	Adrien Gallouët	ggml : fix avx512bf16 build (llama/18623)	commit \| commitdiff \| tree
2026-01-11	Raul Torres	CANN: Make `valid_values` variable `static const` ...	commit \| commitdiff \| tree
2026-01-11	nwyin	ggml webgpu: add CEIL operation support (llama/18605)	commit \| commitdiff \| tree
2026-01-11	Johannes Gäßler	CUDA: fix FA FP16 accumulator overflow for Granite...	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	ggml-cuda: check for srcs outside the cgraph (llama...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: fix topk_moe_sigmoid_norm_bias failures in...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: handle quantize_q8_1 overflowing the max workgr...	commit \| commitdiff \| tree
2026-01-11	Chenguang Li	CANN: add operator fusion support for ADD + RMS_NORM...	commit \| commitdiff \| tree
2026-01-11	Daniel Bevenius	sampling : add support for backend sampling (llama...	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	CUDA: disable cuda graph when using n-cpu-moe (llama...	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	ggml-cuda: remove unused params in ggml_cuda_graph...	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	ggml-cuda: fixes for concurrent streams (llama/18496)	commit \| commitdiff \| tree
2026-01-11	Johannes Gäßler	CUDA: only allocate FA tmp buffer if needed (llama...	commit \| commitdiff \| tree
2026-01-11	pl752	(Bugfix, ggml-cuda) Pool alloc count fix + small size...	commit \| commitdiff \| tree
2026-01-11	Shouyu	ggml-hexagon: optimize activation function (llama/18393)	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: Optimize GGML_OP_CUMSUM (llama/18417)	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: Implement mmvq for iq1_s/iq1_m (llama/18450)	commit \| commitdiff \| tree
2026-01-11	Georgi Gerganov	metal : adjust extra size for FA buffer to avoid reallo...	commit \| commitdiff \| tree
2026-01-11	Chris Rohlf	rpc : use unordered_map::reserve and emplace (llama...	commit \| commitdiff \| tree
2026-01-11	MeeMin	cuda : fix copy of large tensors (ggml_nbytes <= INT_MA...	commit \| commitdiff \| tree
2026-01-11	Aman Gupta	ggml-cuda: remove unneccesary prints on ggml_cuda_init...	commit \| commitdiff \| tree
2026-01-11	Jeff Bolz	vulkan: extend topk_moe to handle sigmoid w/exp_probs_b...	commit \| commitdiff \| tree
2025-12-31	Georgi Gerganov	ggml : bump version to 0.9.5 (#1410) upstream/0.9.5 v0.9.5	commit \| commitdiff \| tree
2025-12-31	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2025-12-31	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-12-31	gatbontonpc	metal : add count_equal op (llama/18314)	commit \| commitdiff \| tree
2025-12-31	Johannes Gäßler	CUDA: fix KQ max calculation (llama/18487)	commit \| commitdiff \| tree
2025-12-31	Georgi Gerganov	metal : remove BF16 x F16 kernels (llama/18456)	commit \| commitdiff \| tree
2025-12-31	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	sycl: add newline at the end of CMakeLists.txt (llama...	commit \| commitdiff \| tree
2025-12-31	Rahul Sathe	Work around broken IntelSYCLConfig.cmake in Intel oneAP...	commit \| commitdiff \| tree
2025-12-31	Charles Xu	kleidiai: add and integrate SVE 256-bit vector-length...	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	CUDA: add log line when mxfp4 acceleration is used...	commit \| commitdiff \| tree
2025-12-31	Johannes Gäßler	CUDA: fix replacment of bad archs in CMake (llama/18457)	commit \| commitdiff \| tree
2025-12-31	Johannes Gäßler	CUDA: Blackwell features for non-native builds (llama...	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	cuda: fix race condition in cumsum (llama/18448)	commit \| commitdiff \| tree
2025-12-31	uvos	HIP: Use mmq on MFMA devices for MUL_MAT_ID in cases...	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	Revert "ggml-cuda: use CMAKE_CUDA_ARCHITECTURES if...	commit \| commitdiff \| tree
2025-12-31	o7si	rpc: fix segfault on invalid endpoint format (llama...	commit \| commitdiff \| tree
2025-12-31	Boian Berberov	cmake: Added more x86_64 CPU backends when building...	commit \| commitdiff \| tree
2025-12-31	QDelta	ggml-cuda: use CMAKE_CUDA_ARCHITECTURES if set when...	commit \| commitdiff \| tree
2025-12-31	lhez	opencl: allow resizing transpose buffers (llama/18384)	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	ggml-cuda: Use same regex for GGML_NATIVE=OFF (llama...	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: preprocess mul_mat_id experts and discard workg...	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: optimize decodeFuncB in coopmat2 mul_mat_id...	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: Use BK=32 for coopmat2 mul_mat_id (llama/18332)	commit \| commitdiff \| tree
2025-12-31	Eve	vulkan: small dequantization improvements (llama/18380)	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: Support UPSCALE w/antialias (llama/18327)	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: handle rope with large number of rows (llama...	commit \| commitdiff \| tree
2025-12-31	0Marble	CANN: implement the SSM_CONV operator (llama/17737)	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	ggml-cuda: fix regex for arch list (llama/18371)	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	cuda: optimize cumsum cub path (llama/18362)	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	ggml-cuda: fix blackwell native builds (llama/18361)	commit \| commitdiff \| tree
2025-12-31	Penglin Cai	CANN: Add support for CONV_TRANSPOSE_1D when kernel...	commit \| commitdiff \| tree
2025-12-31	Aadeshveer...	ggml : optimize cuda cumsum fallback kernel (llama...	commit \| commitdiff \| tree
2025-12-31	Aman Gupta	CUDA: experimental native mxfp4 support for blackwell...	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: fix command buffer corruption in ggml_backend_v...	commit \| commitdiff \| tree
2025-12-31	Wang Weixuan	CANN : refactor ACL graph cache (llama/17752)	commit \| commitdiff \| tree
2025-12-31	Ruben Ortlam	vulkan: use fewer FA rows for small cache runs (llama...	commit \| commitdiff \| tree
2025-12-31	TianHao324	CANN: Uses yarn_ramp cache in ROPE (llama/17725)	commit \| commitdiff \| tree
2025-12-31	Chris Rohlf	rpc : add check for rpc buffer type (llama/18242)	commit \| commitdiff \| tree
2025-12-31	nullname	ggml-hexagon: create generalized functions for cpu...	commit \| commitdiff \| tree
2025-12-31	Shouyu	ggml-hexagon: gelu optimization (llama/18151)	commit \| commitdiff \| tree
2025-12-31	Taimur Ahmad	llamafile: add rvv support for sgemm kernels (llama...	commit \| commitdiff \| tree
2025-12-31	lhez	opencl: unpack q4_0 for adreno in get_tensor (llama...	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: Extend rope fusions to allow mrope (llama/18264)	commit \| commitdiff \| tree
2025-12-31	Jeff Bolz	vulkan: Implement set_tensor_async and the event interf...	commit \| commitdiff \| tree
2025-12-31	Johannes Gäßler	llama: fix RPC for -fit on (llama/18233)	commit \| commitdiff \| tree
next

Packaging of ggml-org/ggml

RSS Atom