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

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

overview / pkg / ggml / sources / ggml / shortlog

2025-03-27	BB-fat	metal : Cache the Metal library at the device context...	commit \| commitdiff \| tree
2025-03-27	Eve	mat vec double buffer (llama/12188)	commit \| commitdiff \| tree
2025-03-27	R0CKSTAR	musa: support new arch mp_31 and update doc (llama...	commit \| commitdiff \| tree
2025-03-27	Henry Linjamäki	opencl: use OpenCL C standard supported by the device...	commit \| commitdiff \| tree
2025-03-27	Georgi Gerganov	tests : fix test-quantize-fns to init the CPU backend...	commit \| commitdiff \| tree
2025-03-27	Jason C.H	ggml-backend : make path_str compatible with C++20...	commit \| commitdiff \| tree
2025-03-27	Daniel Bevenius	ggml : skip intermediate .air file when compiling ...	commit \| commitdiff \| tree
2025-03-26	Akarshan Biswas	ci: disable test-opt for now (#1158)	commit \| commitdiff \| tree
2025-03-25	Akarshan Biswas	ci: Initial SYCL setup (#1157)	commit \| commitdiff \| tree
2025-03-13	cmdr2	Create CONTRIBUTING.md (#1146)	commit \| commitdiff \| tree
2025-03-13	bssrdf	gpt-2 : add comment about KV cache type (#1142)	commit \| commitdiff \| tree
2025-03-10	Christian Kastner	cmake: Enable specifying exact PowerPC CPU architecture...	commit \| commitdiff \| tree
2025-03-10	Christian Kastner	cmake: Comment out GGML_BIN_DIR for now (#1139)	commit \| commitdiff \| tree
2025-03-08	Georgi Gerganov	spm : remove (#1135)	commit \| commitdiff \| tree
2025-03-08	Georgi Gerganov	sync : whisper.cpp upstream/0.0.1802	commit \| commitdiff \| tree
2025-03-08	Dmitry Atamanov	common : fix audio loading by miniaudio (whisper/2862)	commit \| commitdiff \| tree
2025-03-07	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-03-07	Rémy O	ggml-cpu: faster AVX2 variant for IQ1_M (llama/12216)	commit \| commitdiff \| tree
2025-03-07	BB-fat	metal : simplify kernel arguments using a struct (...	commit \| commitdiff \| tree
2025-03-07	Daniel Bevenius	metal : fix default.metallib build (llama/12224)	commit \| commitdiff \| tree
2025-03-07	lhez	opencl: Noncontiguous `norm`, `rms_norm`, disable ...	commit \| commitdiff \| tree
2025-03-07	xiaofei	cmake : fix undefined reference errors for std::filesys...	commit \| commitdiff \| tree
2025-03-07	Johannes Gäßler	CUDA: fix FA logic for PTX 7.0 and CC >= 7.5 (llama...	commit \| commitdiff \| tree
2025-03-07	uvos	HIP/CUDA: set the paramerter value in maintain_cuda_gra...	commit \| commitdiff \| tree
2025-03-07	Henry Linjamäki	opencl : fix buffer alignment (llama/12197)	commit \| commitdiff \| tree
2025-03-07	Henry Linjamäki	opencl : fix `ulong` kernel args were set from `int...	commit \| commitdiff \| tree
2025-03-07	simon886212	opencl : fix profile-related errors (llama/12095)	commit \| commitdiff \| tree
2025-03-07	Rémy O	ggml-cpu: Faster IQ1 mul_mat_vec on AVX2 using BMI2...	commit \| commitdiff \| tree
2025-03-07	Akarshan Biswas	SYCL: Disable f16 Unary OPs as not supported by the...	commit \| commitdiff \| tree
2025-03-07	Plamen Minev	ggml : fix GGMLMetalClass ODR (llama/12200)	commit \| commitdiff \| tree
2025-03-07	vmobilis	ggml : ggml_compute_forward_concat() for arbitrary...	commit \| commitdiff \| tree
2025-03-06	Christian Kastner	pkg-config: Use CMake install paths for lib, include...	commit \| commitdiff \| tree
2025-03-04	Georgi Gerganov	vulkan : sync (llama/0)	commit \| commitdiff \| tree
2025-03-04	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-03-04	mgroeber9110	ggml : portability fixes for VS 2017 (llama/12150)	commit \| commitdiff \| tree
2025-03-04	David Huang	HIP: implement FlashAttention via rocWMMA for CDNA...	commit \| commitdiff \| tree
2025-03-04	Diego Devesa	test-backend-ops : add option -p to filter by op params...	commit \| commitdiff \| tree
2025-03-04	ag2s20150909	ggml : fix kleidiai build (llama/12159)	commit \| commitdiff \| tree
2025-03-04	Akarshan Biswas	SYCL: Move CPY kernels to a separate file and add few...	commit \| commitdiff \| tree
2025-03-04	Diego Devesa	ggml-backend : keep paths in native string type when...	commit \| commitdiff \| tree
2025-03-04	Erik Scholz	CUDA: compress mode option and default to size (llama...	commit \| commitdiff \| tree
2025-03-04	William Tambellini	ggml : upgrade init_tensor API to return a ggml_status...	commit \| commitdiff \| tree
2025-03-04	Rémy O	vulkan: add specific MMV kernels for IQ2 and IQ3 quants...	commit \| commitdiff \| tree
2025-03-04	Johannes Gäßler	CUDA: fix logic for V100 + GGML_CUDA_FORCE_MMQ (llama...	commit \| commitdiff \| tree
2025-03-04	Prashant Vithule	ggml: aarch64: implement SVE kernels for q2_k_q8_k...	commit \| commitdiff \| tree
2025-03-04	hipudding	CANN: Fix build error with GCC 13 (llama/11990)	commit \| commitdiff \| tree
2025-03-04	Eve	vulkan: matmul dequantization improvements (llama/12015)	commit \| commitdiff \| tree
2025-03-04	Daniele	vulkan: improve im2col (llama/11826)	commit \| commitdiff \| tree
2025-03-04	Vladimir Vuksanovic	cmake: Fix ggml backend dependencies and installation...	commit \| commitdiff \| tree
2025-03-04	Jeff Bolz	vulkan: fix assertion when qy_needs_dequant (llama...	commit \| commitdiff \| tree
2025-03-04	Molly Sophia	ggml-cpu: Fix build with sve (llama/12059)	commit \| commitdiff \| tree
2025-03-03	cmdr2	cuda: unary ops as float + de-duplicate (#1130)	commit \| commitdiff \| tree
2025-02-28	cmdr2	cuda/vulkan: specify fp32-only support for some operati...	commit \| commitdiff \| tree
2025-02-28	cmdr2	cuda/cpu: Increase support for fp16 unary operations...	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2025-02-27	Diego Devesa	whisper : support GGML_BACKEND_DL (whisper/2843)	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	ci : fix workflow name	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	examples : remove dr_wab.h (#1127)	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	common : separate whisper sources (whisper/2846)	commit \| commitdiff \| tree
2025-02-27	Georgi Gerganov	common : fix build min/max (whisper/2845)	commit \| commitdiff \| tree
2025-02-27	Dmitry Atamanov	examples : use miniaudio for direct decoding flac,...	commit \| commitdiff \| tree
2025-02-27	midnight	cmake : fix compile assumptions for power9/etc (whisper...	commit \| commitdiff \| tree
2025-02-26	petterreinholdtsen	Told cmake to install ggml-cpp.h as a public header...	commit \| commitdiff \| tree
2025-02-25	cmdr2	Support pure float16 add/sub/mul/div operations in...	commit \| commitdiff \| tree
2025-02-25	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-02-25	Gian-Carlo...	metal : copy kernels for quant to F32/F16 conversions...	commit \| commitdiff \| tree
2025-02-25	lhez	opencl: fix for small models (llama/11950)	commit \| commitdiff \| tree
2025-02-25	Neo Zhang Jianyu	Optimize mul_mat for Q4_0 on Intel GPU (llama/12035)	commit \| commitdiff \| tree
2025-02-25	Akarshan Biswas	SYCL: Fix GGML_SYCL_DEBUG macro (llama/11995)	commit \| commitdiff \| tree
2025-02-25	Aaron Teo	ggml-cpu: Support s390x SIMD Instruction Set (llama...	commit \| commitdiff \| tree
2025-02-25	Johannes Gäßler	CUDA: app option to compile without FlashAttention...	commit \| commitdiff \| tree
2025-02-25	Johannes Gäßler	CUDA: optimize FA for GQA + large batches (llama/12014)	commit \| commitdiff \| tree
2025-02-25	Gian-Carlo...	cuda: Add Q5_1, Q5_0, Q4_1 and Q4_0 to F32 conversion...	commit \| commitdiff \| tree
2025-02-25	PureJourney	CUDA: correct the lowest Maxwell supported by CUDA...	commit \| commitdiff \| tree
2025-02-25	Bodhi	MUSA: support ARM64 and enable dp4a .etc (llama/11843)	commit \| commitdiff \| tree
2025-02-25	Charles Xu	ggml-cpu: Add CPU backend support for KleidiAI library...	commit \| commitdiff \| tree
2025-02-25	Prashant Vithule	ggml: aarch64: implement SVE kernels for q3_K_q8_K...	commit \| commitdiff \| tree
2025-02-25	Johannes Gäßler	CUDA: use async data loading for FlashAttention (llama...	commit \| commitdiff \| tree
2025-02-25	Rémy O	vulkan: implement several ops relevant for ggml_opt...	commit \| commitdiff \| tree
2025-02-25	Jeff Bolz	vulkan: support multi/vision rope, and noncontiguous...	commit \| commitdiff \| tree
2025-02-25	Hale Chan	metal : fix the crash caused by the lack of residency...	commit \| commitdiff \| tree
2025-02-25	Adrian Kretz	metal : optimize dequant q6_K kernel (llama/11892)	commit \| commitdiff \| tree
2025-02-25	Georgi Gerganov	repo : update links to new url (llama/11886)	commit \| commitdiff \| tree
2025-02-25	Rémy O	vulkan: initial support for IQ1_S and IQ1_M quantizatio...	commit \| commitdiff \| tree
2025-02-25	lhez	opencl: Fix rope and softmax (llama/11833)	commit \| commitdiff \| tree
2025-02-25	Diego Devesa	cuda : add ampere to the list of default architectures...	commit \| commitdiff \| tree
2025-02-25	Jinyang He	ggml: optimize some vec dot functions for LoongArch...	commit \| commitdiff \| tree
2025-02-25	Eve	vulkan: linux builds + small subgroup size fixes (llama...	commit \| commitdiff \| tree
2025-02-25	Jeffrey Morgan	llamafile: use member variable instead of constant...	commit \| commitdiff \| tree
2025-02-25	R0CKSTAR	musa: bump MUSA SDK version to rc3.1.1 (llama/11822)	commit \| commitdiff \| tree
2025-02-25	Diego Devesa	ggml-cpu : add chunking support to mul_mat_id (llama...	commit \| commitdiff \| tree
2025-02-25	Xuan-Son Nguyen	ggml : x2 speed for WASM by optimizing SIMD (llama...	commit \| commitdiff \| tree
2025-02-25	uvos	HIP: Remove GCN from list of devices that avoid MMQ...	commit \| commitdiff \| tree
2025-02-12	Georgi Gerganov	sync : llama.cpp upstream/0.0.1722	commit \| commitdiff \| tree
2025-02-12	uvos	HIP: Switch to std::vector in rocblas version check...	commit \| commitdiff \| tree
2025-02-12	bandoti	cleanup: fix compile warnings associated with gnu_print...	commit \| commitdiff \| tree
2025-02-12	Richard	ggml : fix multi-threaded clamp_f32 (llama/11824)	commit \| commitdiff \| tree
2025-02-12	Weizhao Ouyang	ggml-cpu: Fix duplicate MATMUL_INT8 (llama/11817)	commit \| commitdiff \| tree
2025-02-12	Johannes Gäßler	CUDA: fix CUDART_VERSION checks (llama/11821)	commit \| commitdiff \| tree
next

Packaging of ggml-org/ggml

RSS Atom