]> git.djapps.eu Git - pkg/ggml/sources/ggml/shortlog
pkg/ggml/sources/ggml
2025-09-05 Jeff Bolzvulkan: Skip syncing for prealloc_y when it is reused...
2025-09-05 Chenguang LiCANN: FIx compiler warnings (llama/15661)
2025-09-05 Aman GuptaCUDA: fix bug in rms_norm fusion (llama/15660)
2025-09-05 Aman GuptaCUDA: fuse adds, fuse add with rms norm (llama/15631)
2025-09-05 mnehete32CUDA: add conv2d (llama/15635)
2025-09-05 Aaron Teoggml-cpu: fix invalid hsum build in debug s390x (llama...
2025-09-05 compiladeggml : fix SSM_SCAN for n_groups > 1 (llama/15625)
2025-09-05 Georgi Gerganovkv-cache : remove LLAMA_SET_ROWS checks (llama/15505)
2025-09-05 matiaslincuda: Add cublasLt_static linking when GGML_STATIC...
2025-09-05 uvosHIP: Enable support for ggml_backend_cuda_register_host...
2025-09-05 Chenguang LiCANN: refactor mask handling and improve performance...
2025-09-05 xctanggml-cpu : add basic RVV support for vector f32 ops...
2025-09-05 rmatifOpenCL: add fused group_norm/norm, mul, add (llama...
2025-09-05 Diego Devesatests : fix test-opt with GGML_BACKEND_DL (llama/15599)
2025-09-05 Akarshan BiswasSYCL: fix rms_norm_mul_add for tensor dim not a multipl...
2025-09-05 Evetests: add performance test for mul mat id (llama/15543)
2025-09-05 shalinib-ibmllamafile: PowerPC Sgemm Optimization (llama/15558)
2025-09-05 Johannes GäßlerCUDA: return -1 for nonexistent compiled arch (llama...
2025-09-05 Georgi Gerganovmetal : optimize FA vec for large sequences and BS...
2025-09-05 Georgi Gerganovmetal : improve `MUL_MAT_ID` (llama/15541)
2025-09-05 Sigbjørn Skjæretmetal : remove contiguous assertion for src0 in IM2COL...
2025-09-05 Yoshi_likes_e4Add a warning for special devices (llama/15563)
2025-09-05 Jeff Bolzvulkan: Remove splitting for mul_mat_id (llama/15568)
2025-09-05 QeeweewCUDA: Accelerate MXFP4 table lookup using `__byte_perm...
2025-09-05 lhezopencl: fix support ops condition for `rms_norm` (llama...
2025-09-05 Ruben Ortlamvulkan: fix min subgroup 16 condition for mmid subgroup...
2025-09-05 Jeff Bolztests: Generate unique input values for count_equal...
2025-09-05 Ihar Hrachyshkametal: fix regression when no metal devices are present...
2025-09-05 Johannes GäßlerCUDA: MoE helper in device code, better tile sizes...
2025-09-05 Georgi Gerganovmetal : add FA kernels for HS=40 (llama/15559)
2025-09-05 Chenguang LiCANN: ROPE cache sin/cos repeat (llama/15501)
2025-09-05 Ruben Ortlamvulkan: apply MUL_MAT_ID subgroup optimization to non...
2025-09-05 Jeff Bolzvulkan: Support FA with any multiple of 8 head sizes...
2025-09-05 Ruben Ortlamvulkan: enable Conv2D for Apple after MoltenVK fixed...
2025-09-05 Jeff Bolzvulkan: workaround MoltenVK compile failure in multi_ad...
2025-09-05 Johannes GäßlerCUDA: fix half2 -> half conversion for HIP (llama/15529)
2025-09-05 Jeff Bolzvulkan: optimize rms_norm, and allow the work to spread...
2025-09-05 Jeff Bolzvulkan: Rewrite synchronization to allow some overlap...
2025-09-05 Aclyvulkan : support ggml_mean (llama/15393)
2025-09-05 Jeff Bolzvulkan: optimize mul_mat_id loading row ids into shared...
2025-09-05 Johannes Gäßlertest-opt: allow slight inprecision (llama/15503)
2025-09-05 Reese Levineggml WebGPU: add support for quantization types (llama...
2025-09-05 rmatifggml: add `conv3d` op (llama/15182)
2025-09-05 Yavor Ivanovcuda : add Pad Reflect 1D support (llama/14659)
2025-09-05 Aaron Teoggml-cpu: Support Q5_0 and Q5_1 on s390x (llama/15486)
2025-09-05 Chenguang LiCANN: Optimize RMS_NORM using cache (llama/15419)
2025-09-05 Diego Devesasched : fix possible use of wrong ids tensor when offlo...
2025-09-05 Aclyvulkan : support conv_2d_dw with f16 weights (llama...
2025-09-05 Dong Won Kimvulkan: add exp operation (llama/15456)
2025-09-05 Jeff Bolzvulkan: Reuse conversion results in prealloc_y (llama...
2025-09-05 Xuan-Son Nguyenggml : fix condition of im2col on Metal backend (llama...
2025-09-05 R0CKSTARmusa: add GGML_UNUSED_VARS (llama/15446)
2025-09-05 Diego Devesasched : copy only the used experts when offloading...
2025-09-05 Johannes GäßlerCUDA: refactor FA support/selection code (llama/15454)
2025-09-05 Johannes GäßlerCUDA: replace GGML_CUDA_F16 with CUDA arch checks ...
2025-09-05 Jeff Bolzvulkan: shorten pipeline name strings (llama/15431)
2025-09-05 R0CKSTARmusa: fix build warnings (llama/15258)
2025-09-05 lhezopencl: mark `argsort` unsupported if cols exceed workg...
2025-09-05 SHUAI YANGCANN: optimize rope operator (llama/15335)
2025-09-05 R0CKSTARmusa: handle __hgt2_mask, available starting from MUSA...
2025-09-05 Marvin Gießingggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le...
2025-08-28 Daniel Beveniusci : add github release job (#1334)
2025-08-18 Georgi Gerganovcuda : remove obsolete sources (#1332) upstream/0.0.2471
2025-08-18 Georgi Gerganovsync : whisper.cpp
2025-08-18 Georgi Gerganovscripts : update sync scripts
2025-08-18 Reese Levineggml: Add initial WebGPU backend (llama/14521)
2025-08-18 Aaron Teoggml : initial zDNN backend (llama/14975)
2025-08-18 Georgi Gerganovscripts : update sync scripts
2025-08-18 Georgi Gerganovcommon : handle mxfp4 enum
2025-08-18 Georgi Gerganovsync : llama.cpp
2025-08-18 compiladeggml-quants : fix make_qp_quants NANs and IQ1 assertion...
2025-08-18 Jeff Bolzvulkan: disable spirv-opt for bfloat16 shaders (llama...
2025-08-18 Jeff Bolzvulkan: Use larger workgroups for mul_mat_vec when...
2025-08-18 Dong Won Kimvulkan: support sqrt (llama/15370)
2025-08-18 Jeff Bolzvulkan: Optimize argsort (llama/15354)
2025-08-18 Jeff Bolzvulkan: fuse adds (llama/15252)
2025-08-18 Jeff Bolzvulkan: Support mul_mat_id with f32 accumulators (llama...
2025-08-18 Jeff Bolzvulkan: Add missing bounds checking to scalar/coopmat1...
2025-08-18 rmatifOpenCL: add initial FA support (llama/14987)
2025-08-18 lhezopencl: add initial mxfp4 support via mv (llama/15270)
2025-08-18 Georgi Gerganovvulkan : fix out-of-bounds access in argmax kernel...
2025-08-18 Georgi Gerganovvulkan : fix compile warnings on macos (llama/15340)
2025-08-18 Aaron Teoggml: initial IBM zDNN backend (llama/14975)
2025-08-18 Johannes Gäßlertest-opt: fix backend support check (llama/15317)
2025-08-18 Johannes GäßlerCUDA: fix negative KV_max values in FA (llama/15321)
2025-08-18 uvosHIP: Cleanup hipification header (llama/15285)
2025-08-18 Jeff Bolzvulkan: perf_logger improvements (llama/15246)
2025-08-14 Jason Niggml: fix ggml_conv_1d_dw bug (#1323) upstream/0.0.2446
2025-08-14 Georgi Gerganovmnist : adapt to opt changes
2025-08-14 Georgi Gerganovtests : remove unused includes (#0)
2025-08-14 Georgi Gerganovsync : llama.cpp
2025-08-14 Sigbjørn Skjæretcuda : fix GGML_CUDA_GRAPHS=OFF (llama/15300)
2025-08-14 Jonathan Graehlfinetune: SGD optimizer, more CLI args (llama/13873)
2025-08-14 uvosHIP: bump requirement to rocm 6.1 (llama/15296)
2025-08-14 Georgi Gerganovsync : llama.cpp
2025-08-14 Juddggml : update `ggml_rope_multi` (llama/12665)
2025-08-14 Georgi Gerganovggml : repack block_iq4_nlx8 (llama/14904)
2025-08-14 Oliver SimonsCUDA: Optimize `reduce_rows_f32` kernel, leading up...
2025-08-14 Tak-RSggml-rpc: chunk send()/recv() to avoid EINVAL for very...
2025-08-14 uvosHIP: disable sync warp shuffel operators from clr amd_w...
next