]> git.djapps.eu Git - pkg/ggml/sources/ggml/shortlog
pkg/ggml/sources/ggml
2025-08-02 AN Longsimple : fix typo (#1319)
2025-08-01 Mathieu BaudierUpdate upstream
2025-08-01 Mathieu BaudierMerge tag 'upstream/0.0.2309' into debian/latest
2025-07-30 Georgi Gerganovsync : whisper.cpp
2025-07-30 Kai Pastorcmake : Fix BLAS link interface (#1316)
2025-07-30 Kai Pastorvulkan : fix 32-bit builds (#1313)
2025-07-28 Georgi Gerganovsync : llama.cpp
2025-07-28 Erik Scholzvulkan : add fp16 support for the conv_2d kernel (llama...
2025-07-28 Jeff Bolzvulkan: skip empty set_rows to avoid invalid API usage...
2025-07-28 Aman GuptaDocs: add instructions for adding backends (llama/14889)
2025-07-28 deepsekHIP: Enable Matrix cores for MMQ Kernels, Enable stream...
2025-07-28 hipuddingCANN: Implement GLU ops (llama/14884)
2025-07-28 R0CKSTARmusa: fix build warnings (unused variable) (llama/14869)
2025-07-28 Aaron Teoggml-cpu : disable GGML_NNPA by default due to instabil...
2025-07-28 Gabe Goodhartmetal: SSM_SCAN performance (llama/14743)
2025-07-28 lhezopencl: add fused `rms_norm_mul` (llama/14841)
2025-07-28 Oliver Simonsggml : remove invalid portPos specifiers from dot files...
2025-07-28 Chris Rohlfrpc : check for null buffers in get/set/copy tensor...
2025-07-28 Diego Devesasched : fix multiple evaluations of the same graph...
2025-07-28 R0CKSTARmusa: upgrade musa sdk to rc4.2.0 (llama/14498)
2025-07-25 Georgi Gerganovcontrib : recommend PRs to llama.cpp (#1312)
2025-07-24 Kai Pastorcmake : Indent ggml-config.cmake (#1310)
2025-07-24 Georgi Gerganovsync : llama.cpp
2025-07-24 Alberto Cabrera... sycl: fixed semantics of block offset calculation ...
2025-07-24 Georgi Gerganovmetal : fix fusion across different encoders (llama...
2025-07-24 Donghyeon Jeongsycl: fix undefined variable in work group size check...
2025-07-24 Johannes GäßlerCUDA: fix overflow in FA, tune performance (llama/14840)
2025-07-24 Johannes GäßlerCUDA: fix compilation with GGML_CUDA_F16 (llama/14837)
2025-07-24 Johannes GäßlerCUDA: fix quantized KV cache + multiple sequences ...
2025-07-24 Georgi Gerganovtests : add non-cont K,V FA tests
2025-07-24 lixing-starggml: fix loongarch quantize_row_q8_1 error (llama...
2025-07-24 chen fanCANN: weight format to NZ for Ascend310P3 (llama/14407)
2025-07-24 Aman GuptaCUDA: add fused rms norm (llama/14800)
2025-07-24 Jeff Bolzvulkan: fix rms_norm_mul to handle broadcasting dim0...
2025-07-24 Sigbjørn Skjæretcuda : implement bf16 cpy ops and enable bf16 cont...
2025-07-24 lhezopencl: remove unreachable `return` (llama/14806)
2025-07-24 R0CKSTARcuda: remove linking to cublasLt (llama/14790)
2025-07-24 Sigbjørn Skjæretopencl: fix `im2col` when `KW!=KH` (llama/14803)
2025-07-24 rmatifopencl: add conv2d kernel (llama/14403)
2025-07-24 Romain Biessysycl: Fix im2col (llama/14797)
2025-07-24 Charles Xukleidiai: add support for get_rows (llama/14676)
2025-07-24 Jeff Bolzvulkan/cuda: Fix im2col when KW!=KH (llama/14789)
2025-07-24 Ervin Áron... ggml: adds CONV_2D op and direct GEMM Vulkan implementa...
2025-07-24 Peter0x44vulkan: Add logging for bf16 features to ggml_vk_print_...
2025-07-24 0cc4mVulkan: Fix fprintf format-security warning (llama...
2025-07-23 Kai PastorCI: Test static build (#1307)
2025-07-22 Kai Pastorcmake : fix usage issues (#1257)
2025-07-21 Daniel Beveniusggml-cpu : remove stdlib include from repack.cpp (...
2025-07-19 Georgi Gerganovsync : llama.cpp
2025-07-19 Georgi Gerganovmetal : fuse add, mul + add tests (llama/14596)
2025-07-19 Oliver Simonscuda : Fix Gemma3n not executed as CUDA_GRAPH on NVGPUs...
2025-07-19 Aman GuptaCUDA: set_rows + cpy.cu refactor (llama/14712)
2025-07-19 Neo Zhang Jianyuuse max work group size for device to replace the magic...
2025-07-19 Reese Levineggml: Add initial WebGPU backend (llama/14521)
2025-07-19 Georgi Gerganovllama : add high-throughput mode (llama/14363)
2025-07-19 Georgi Gerganovggml : add asserts (llama/14720)
2025-07-19 Jeff Bolzvulkan: fix noncontig check for mat_mul_id splitting...
2025-07-19 Jeff Bolzvulkan: add RTE variants for glu/add/sub/mul/div (llama...
2025-07-19 R0CKSTARcuda: fix build warnings in set-rows.cu (unused variabl...
2025-07-19 Anton Mitkovsycl: Hotfix for non dnnl codepath (llama/14677)
2025-07-19 shalinib-ibmggml : refactor llamafile_sgemm PPC code (llama/14673)
2025-07-19 Akarshan BiswasSYCL: use 1D kernel for set_rows (llama/14618)
2025-07-19 Anton Mitkovsycl: Batched mulmat rework for oneDNN dispatch (llama...
2025-07-19 Sigbjørn Skjæretcuda : add set rows for bf16 (llama/14664)
2025-07-19 Yavor Ivanovcuda : add ELU support (llama/14657)
2025-07-19 Georgi Gerganovggml : add build-time message to remind about ggml_set_...
2025-07-19 Yavor Ivanovmetal : Add missing unary ops Metal support (llama...
2025-07-19 Tarek Dakhrantests : cover lfm2 cases in test_ssm_conv (llama/14651)
2025-07-19 Aman GuptaCUDA: add set rows for f32 and f16 (llama/14551)
2025-07-12 Georgi Gerganovsync : whispper.cpp
2025-07-12 Georgi Gerganovgit : remove kompute submodule (#1300) upstream/0.0.2309
2025-07-12 Georgi Gerganovsync : resolve conflicts (#0)
2025-07-12 Georgi Gerganovsync : llama.cpp
2025-07-12 Jeff Bolzvulkan: support SET_ROWS (llama/14587)
2025-07-12 Jeff Bolzvulkan: optimizations for deepseek prompt processing...
2025-07-12 Tarek Dakhranmodel : support LiquidAI LFM2 hybrid family (llama...
2025-07-12 Slobodan JosicHIP : Add HIP 7.0+ compatibility for hipBLAS compute...
2025-07-12 rmatifopencl: add tiled mul_mat_f16_f32 (llama/14535)
2025-07-12 lhezopencl: add `set_rows` for `f16` and `f32` (llama/14547)
2025-07-12 Aman GuptaDocs: script to auto-generate ggml operations docs...
2025-07-12 Akarshan BiswasSYCL: Initial set_rows kernel implementation (llama...
2025-07-12 compiladecuda : support Falcon-H1 state size for SSM_SCAN (llama...
2025-07-12 Xuan-Son Nguyenggml : add ggml_scale_bias (llama/14417)
2025-07-12 Miaoqian Linggml : prevent integer overflow in gguf tensor size...
2025-07-12 Jeff Bolzvulkan: optimize flash attention split_k_reduce (llama...
2025-07-12 Jeff Bolzvulkan : fix rope with partial rotation and non-cont...
2025-07-12 Georgi Gerganovcuda : fix rope with partial rotation and non-cont...
2025-07-12 Aman GuptaCUDA: add bilinear interpolation for upscale (llama...
2025-07-12 R0CKSTARmusa: fix build warnings (unused variable) (llama/14561)
2025-07-12 Aman GuptaCUDA: add bf16 and i32 to getrows (llama/14529)
2025-07-12 Evevulkan: increase LOAD_VEC_A to 8 (IQ1/IQ2) or 4 (IQ3...
2025-07-12 Jeff Bolzvulkan: fix rms_norm+mul fusion (llama/14545)
2025-07-12 Jeff Bolzvulkan: Handle updated FA dim2/3 definition (llama...
2025-07-12 Sigbjørn Skjæretopencl: add GELU_ERF (llama/14476)
2025-07-12 R0CKSTARtest-backend-ops: add support for specifying output...
2025-07-12 Georgi Gerganovmetal : disable fast math in all quantize kernels ...
2025-07-12 luyhcsuCANN: Replace aclrtMemsetSync with aclnnInplaceZero...
2025-07-12 Sigbjørn Skjæretggml : implement GEGLU_ERF and GEGLU_QUICK ops (llama...
2025-07-12 lhezopencl : broadcast for soft_max (llama/14510)
2025-07-12 Jeff Bolzvulkan: support mixed/deepseekR1 FA head sizes (llama...
next