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

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

overview / pkg / ggml / sources / ggml / shortlog

2025-11-09	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-11-09	Ruben Ortlam	vulkan: iGPU memory reporting fix (llama/17110)	commit \| commitdiff \| tree
2025-11-09	Ruben Ortlam	vulkan: fix mmq out of bounds reads (llama/17108)	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: fuse mul_mat_id + mul (llama/17095)	commit \| commitdiff \| tree
2025-11-09	Georgi Gerganov	metal : retain src and dst buffers during async ops...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: Use spec constants for conv2d s/d/p and kernel...	commit \| commitdiff \| tree
2025-11-09	Aman Gupta	Revert "CUDA: add expert reduce kernel (#16857)" (llama...	commit \| commitdiff \| tree
2025-11-09	Aman Gupta	CUDA: skip fusion for repeating adds in bias (llama...	commit \| commitdiff \| tree
2025-11-09	SavicStefan	vulkan: Increase BK to 32; use BK/4 for non-CM mul_mm...	commit \| commitdiff \| tree
2025-11-09	Aleksei Nikiforov	ggml: disable vxe for cross-compilation by default...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: fuse rms_norm + mul + rope (+ view + set_rows...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: Fix test-thread-safety crashes (llama/17024)	commit \| commitdiff \| tree
2025-11-09	Johannes Gäßler	CUDA: fix MMQ stream-k fixup ne1 indices (llama/17089)	commit \| commitdiff \| tree
2025-11-09	Reese Levine	ggml webgpu: faster matrix multiplication/matrix-vector...	commit \| commitdiff \| tree
2025-11-09	bssrdf	CUDA: properly handle nb00=nb02 case for cpy (llama...	commit \| commitdiff \| tree
2025-11-09	Acly	vulkan : refactor buffer handling in vk_op_f32 (llama...	commit \| commitdiff \| tree
2025-11-09	Johannes Gäßler	CUDA: fix should_use_mmvf for ne11 == 1 (llama/17085)	commit \| commitdiff \| tree
2025-11-09	Adrien Gallouët	Revert "ggml-cpu: detect correct cpu flags for arm64...	commit \| commitdiff \| tree
2025-11-09	iron	ggml-cpu: detect correct cpu flags for arm64 (#16229...	commit \| commitdiff \| tree
2025-11-09	xctan	ggml-cpu : optimize RVV q2_k and q3_k kernels (llama...	commit \| commitdiff \| tree
2025-11-09	Johannes Gäßler	CUDA: fix crash on uneven context without FA (llama...	commit \| commitdiff \| tree
2025-11-09	Georgi Gerganov	metal : initial Metal4 tensor API support (llama/16634)	commit \| commitdiff \| tree
2025-11-09	YehuditE	sycl: add CONCAT operator support (llama/16047)	commit \| commitdiff \| tree
2025-11-09	l3utterfly	ggml-hexagon: graceful fallback for older socs where...	commit \| commitdiff \| tree
2025-11-09	bssrdf	improve CUDA cpy memory bandwidth when copying transpos...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: Fix GGML_VULKAN_CHECK_RESULTS to better handle...	commit \| commitdiff \| tree
2025-11-09	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-11-09	Reese Levine	ggml webgpu: minor set rows optimization (llama/16810)	commit \| commitdiff \| tree
2025-11-09	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-11-09	nullname	refactor: replace sprintf with snprintf for safer strin...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: remove the need for the dryrun (llama/16826)	commit \| commitdiff \| tree
2025-11-09	Acly	ggml-cpu : bicubic interpolation (llama/16891)	commit \| commitdiff \| tree
2025-11-09	Noah	Fix garbled output with REPACK at high thread counts...	commit \| commitdiff \| tree
2025-11-09	Aman Gupta	CUDA: avoid mul + bias fusion when doing fusion (llama...	commit \| commitdiff \| tree
2025-11-09	lhez	opencl: support imrope (llama/16914)	commit \| commitdiff \| tree
2025-11-09	theo77186	ggml: CUDA: add head size 72 for flash-attn (llama...	commit \| commitdiff \| tree
2025-11-09	Jinyang He	ggml : LoongArch fixes (llama/16958)	commit \| commitdiff \| tree
2025-11-09	shani-f	SYCL: optimized repeat_back kernel (3× fewer asm instru...	commit \| commitdiff \| tree
2025-11-09	Shagun Bera	test-backend-ops : fix segfault in moe-expert-reduce...	commit \| commitdiff \| tree
2025-11-09	Georgi Gerganov	clip : use FA (llama/16837)	commit \| commitdiff \| tree
2025-11-09	mnehete32	CUDA: add FLOOR, CEIL, ROUND, TRUNC unary ops (llama...	commit \| commitdiff \| tree
2025-11-09	Aaron Teo	ggml: add s390x cpu-feats (llama/16774)	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: Fix multi_add invalid descriptor usage (llama...	commit \| commitdiff \| tree
2025-11-09	Jeff Bolz	vulkan: fuse mul_mat+add and mul_mat_id+add_id (llama...	commit \| commitdiff \| tree
2025-11-09	Oliver Simons	CUDA: Remove unneded bias/gate dims in fused mmvq ...	commit \| commitdiff \| tree
2025-11-09	Johannes Gäßler	CUDA: Volta tensor core support for MMF (llama/16843)	commit \| commitdiff \| tree
2025-11-04	Georgi Gerganov	ggml : fix conv2d_dw SVE path (#1380)	commit \| commitdiff \| tree
2025-11-01	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: add expert reduce kernel (llama/16857)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: disable spirv-opt for rope shaders (llama/16872)	commit \| commitdiff \| tree
2025-11-01	Masato Nakasaka	vulkan: Fix crash when FP16 mul_mat accumulation is...	commit \| commitdiff \| tree
2025-11-01	Ruben Ortlam	vulkan: fix shmem overrun in mmq id shader (llama/16873)	commit \| commitdiff \| tree
2025-11-01	l3utterfly	ggml-hexagon: respect input size when getting/setting...	commit \| commitdiff \| tree
2025-11-01	lhez	opencl: fix boundary handling for mul_mm (llama/16875)	commit \| commitdiff \| tree
2025-11-01	Max Krasnyansky	cpu: introduce chunking for repack matmuls and enable...	commit \| commitdiff \| tree
2025-11-01	JJJYmmm	model: add support for qwen3vl series (llama/16780)	commit \| commitdiff \| tree
2025-11-01	Max Krasnyansky	cpu: introduce chunking for flash attention (llama...	commit \| commitdiff \| tree
2025-11-01	Sigbjørn Skjæret	cuda : fix argsort with 64k+ rows (llama/16849)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: Handle argsort with a large number of rows...	commit \| commitdiff \| tree
2025-11-01	Oliver Simons	Hide latency of bias and gate-loading (llama/16847)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: Fuse rope+set_rows (llama/16769)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: Update topk_moe fusion to handle gpt's late...	commit \| commitdiff \| tree
2025-11-01	Ruben Ortlam	Vulkan MMQ Integer Dot Refactor and K-Quant support...	commit \| commitdiff \| tree
2025-11-01	Max Krasnyansky	Hexagon Op queue & dispatch optimizations (llama/16820)	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: use fastdiv in set-rows (llama/16834)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: Call ggml_vk_buffer_write_2d from ggml_vk_buffe...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: Fix bug in topk-moe for gpt-oss (llama/16821)	commit \| commitdiff \| tree
2025-11-01	YaelLogic	sycl: add RMS_NORM_BACK operation support (llama/16808)	commit \| commitdiff \| tree
2025-11-01	YaelGitAccount	cuda: add SET operation support (llama/16804)	commit \| commitdiff \| tree
2025-11-01	l3utterfly	initialise buffer.device in ggml_hexagon_session (llama...	commit \| commitdiff \| tree
2025-11-01	Chenguang Li	CANN: Improve device ID handling and aclnnArange checks...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: add unused vars to mmvf and mmvq (llama/16807)	commit \| commitdiff \| tree
2025-11-01	tamarPal	sycl: add SSM_CONV operation support (llama/16800)	commit \| commitdiff \| tree
2025-11-01	Acly	ggml : fix interpolate with align-corners and ne=1...	commit \| commitdiff \| tree
2025-11-01	Johannes Gäßler	HIP: fix AMDGPU_TARGETS, update documentation (llama...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	test-backend-ops: print failed tests at the end (llama...	commit \| commitdiff \| tree
2025-11-01	tamarPal	sycl: add ROLL operation support (llama/16665)	commit \| commitdiff \| tree
2025-11-01	shani-f	sycl: add REPEAT_BACK operation support (llama/16734)	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: support for weight clamp in top-k norm (llama...	commit \| commitdiff \| tree
2025-11-01	Acly	ggml-alloc : make gallocr prefer chunks that allow...	commit \| commitdiff \| tree
2025-11-01	Sigbjørn Skjæret	cuda : use fast copy when src and dst are of different...	commit \| commitdiff \| tree
2025-11-01	leejet	ggml: fix cuda kernel launch configuration for k_comput...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: General GEMV fusion (llama/16715)	commit \| commitdiff \| tree
2025-11-01	Gilad S.	vulkan: deduplicate Microsoft Direct3D12 devices (llama...	commit \| commitdiff \| tree
2025-11-01	Giuseppe Scrivano	vulkan: delete dead code (llama/16732)	commit \| commitdiff \| tree
2025-11-01	Jeff Bolz	vulkan: Optimize SSM_SCAN (llama/16645)	commit \| commitdiff \| tree
2025-11-01	leejet	ggml: fix CUDA grid launch condition for large block_nu...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: use CUB for arbitary size argsort (llama/16754)	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	ggml-cuda: use passed ops instead of hardcoded ops...	commit \| commitdiff \| tree
2025-11-01	Matthew Michel	sycl: use async memory allocation to fix crashes during...	commit \| commitdiff \| tree
2025-11-01	Max Krasnyansky	Add experimental ggml-hexagon backend for the Hexagon...	commit \| commitdiff \| tree
2025-11-01	Diego Devesa	Revert "ggml : Leverage the existing GGML_F32_VEC helpe...	commit \| commitdiff \| tree
2025-11-01	sirus20x6	ggml : Leverage the existing GGML_F32_VEC helpers to...	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: fix bug in topk-moe softmax (llama/16711)	commit \| commitdiff \| tree
2025-11-01	Aman Gupta	CUDA: topk-moe: add optional parameter for gpt-oss...	commit \| commitdiff \| tree
2025-11-01	Johannes Gäßler	CUDA: better error for FA kernel with 0 occupancy ...	commit \| commitdiff \| tree
2025-10-29	Jeff Bolz	Rewrite simple-backend to use sched and ggml_backend_lo...	commit \| commitdiff \| tree
2025-10-22	Georgi Gerganov	sync : whisper.cpp	commit \| commitdiff \| tree
2025-10-21	Georgi Gerganov	sync : llama.cpp	commit \| commitdiff \| tree
2025-10-21	Aman Gupta	ggml: add ggml_can_fuse_subgraph (llama/16662)	commit \| commitdiff \| tree
next

Packaging of ggml-org/ggml

RSS Atom