]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2025-09-22 Gabe Goodhartfeat: Add conversion support in GraniteHybrid for non...
2025-09-22 Haiyue Wangclang-tidy : disable warning about performance enum...
2025-09-22 Sigbjørn Skjæretggml : implement set_rows with i32 index (#16159)
2025-09-22 Georgi Gerganovcodeowners : update + cleanup (#16174)
2025-09-22 Adrien Gallouëtcommon : enable `--offline` mode without curl support...
2025-09-22 Quentin Bramaswebui : fix handling incomplete chunks (#16107)
2025-09-22 GideonSerfembedding : fix typos in README (#16171)
2025-09-22 Haiyue Wangcommon : remove unused local variables (#16140)
2025-09-22 Georgi Gerganovggml : extend ggml_can_fuse to work with non-sequential...
2025-09-22 Georgi Gerganovggml : add ggml_op_is_empty (#16122)
2025-09-22 Xuan-Son Nguyencodeowners : update ownership for @ngxson and @allozuar...
2025-09-22 Shin-myoung... Vulkan: add conv_transpose_2d operation (#16022)
2025-09-22 Sigbjørn Skjæretcodeowners : claim responsibility for ci, models, gguf...
2025-09-22 Georgi Gerganovcontrib : update roles (#16113)
2025-09-22 Georgi Gerganovci : remove vulkaninfo calls (#16169)
2025-09-22 Georgi Gerganovci : use smaller model (#16168)
2025-09-22 Jeff Bolzvulkan: add RTE variants of exp shader (#16165)
2025-09-22 Georgi Gerganovci : adjust params for less runtime (#16167)
2025-09-22 Ruben Ortlamvulkan: vec dot matrix multiplication fix (#16151)
2025-09-21 lhezopencl: fix concat crash on win arm64 with Adreno ...
2025-09-21 lhezopencl: initial `q8_0` mv support (#15732)
2025-09-21 Georgi Gerganovci : add label for the RISC-V runner (#16150)
2025-09-21 Georgi Gerganovci : migrate ggml ci to self-hosted runners (#16116)
2025-09-21 Giuseppe Scrivanovulkan: optimize UMA buffer operations and fix driver...
2025-09-21 Jeff Bolzvulkan: fix validation error about VK_PIPELINE_CREATE_C...
2025-09-20 Georgi Gerganovsync : ggml upstream/0.0.6527
2025-09-20 Daniel Beveniusggml : introduce semantic versioning (ggml/1336)
2025-09-20 Gregor JasnyCUDA : conditionally add cuda architectures (ggml/1341)
2025-09-20 Ruben Ortlamvulkan: use vec dot for matrix matrix multiplications...
2025-09-20 Benniserver: fix SSE and OpenAI compatibility for error...
2025-09-19 ssweensllama-bench: add --devices and --list-devices support...
2025-09-19 shun095chat: Fix streaming parser for granite models (#15682)
2025-09-19 Aleksander... feat: Improve mobile UI for Settings Dialog (#16084)
2025-09-19 Xuan-Son Nguyenchat : fix build on arm64 (#16101)
2025-09-19 Xuan-Son Nguyenggml : refactor forward_dup for cpu backend (#16062)
2025-09-18 Adrien Gallouëtggml-amx : fix ggml_amx_init() on generic Linux (#16049)
2025-09-18 Adrien Gallouëtcmake : fix static linking for OpenMP on Unix-like...
2025-09-18 Shawn Guopencl: optimize mxfp4 kernels (#16037)
2025-09-18 Jeff Bolzrename optimize_graph to graph_optimize (#16082)
2025-09-18 Bowen HanCUDA: Optimize PAD_REFLECT_1D (#15957)
2025-09-18 Johannes GäßlerCUDA: fix compilation on CC 6.0 (#16091)
2025-09-18 Eric CurtinAdd resumable downloads for llama-server model loading...
2025-09-18 Georgi Gerganovmetal : use function constants for mul_mv_ext kernels...
2025-09-18 Sigbjørn Skjæretcuda : add missing F32<->I32 entries in ggml_cuda_cpy_f...
2025-09-18 Radoslav Gerganovserver : include usage statistics only when user reques...
2025-09-18 Georgi Gerganovllama : bump max seq limit from 64 to 256 (#15916)
2025-09-18 Georgi Gerganovmetal : improve F32, F16 and BF16 mat-vec multiplicatio...
2025-09-18 Jhen-Jie Hongmetal : avoid call free for non-owned buffer (#16067)
2025-09-18 Georgi Gerganovmetal : handle nil cv during pipeline creation (#16065)
2025-09-18 Chenguang LiCANN: Remove print (#16044)
2025-09-17 Reese LevineGGML WebGPU: Support for ADD, MUL, RMS_NORM, GET_ROWS...
2025-09-17 Georgi Gerganovmetal : refactor + optimize v2 (#15995)
2025-09-17 Aleksander... SvelteKit-based WebUI (#14839)
2025-09-17 Xuan-Son Nguyenconvert : add Llama4ForCausalLM (#16042)
2025-09-17 Johannes GäßlerCUDA: fix FA occupancy, optimize tile kernel (#15982)
2025-09-17 David Ribeiro... common : Fix corrupted memory error on json grammar...
2025-09-17 Evevulkan: automatically remove unsupported devices (...
2025-09-17 Daniel Beveniusci : revert back to macos-13 for macOS-latest-cmake...
2025-09-17 Jie Fu (傅杰)llama-quant : fix the verification of attention layers...
2025-09-17 Jie Fu (傅杰)examples : support encoder-decoder models in the simple...
2025-09-17 Shane Amodel : add OLMo3 support (#16015)
2025-09-17 Chenguang LiCANN: Optimize ggml_cann_set_device (#15935)
2025-09-16 jacekpoplawskillama-bench: add --n-cpu-moe support (#15952)
2025-09-16 Daniel Beveniusci : use macos-latest for arm64 webgpu build (#16029)
2025-09-16 Daniel Beveniusggml : fix padding in timestep embedding kernels (...
2025-09-16 Daniel Beveniusci : upload xcframework artifact from ios-xcode-build...
2025-09-16 Bowen Hanfix: apply clang-format to CUDA macros (#16017)
2025-09-16 Daniel Beveniusci : update macos-latest* jobs to use macos-latest...
2025-09-16 Yuri Khrustalevcmake : Do not install tools on iOS targets (#15903)
2025-09-16 Aman GuptaAdd LLaDA-7b-MoE diffusion model (#16003)
2025-09-15 Jake KarnesCUDA: fix im2col_3d to respect non-contiguous inputs...
2025-09-15 Diego Devesadocker : enable rocWMMA in ROCm images, add gfx1151...
2025-09-15 Diego Devesareleases : switch to rocWMMA develop branch, add gfx115...
2025-09-15 yael-worksSYCL: Add COUNT_EQUAL operator support (#15991)
2025-09-15 Nikolay Popovllama-run: Fix model download on Windows (#15988)
2025-09-15 Aman GuptaCUDA: some micro-optimizations in mmf.cuh for mul_mat_i...
2025-09-15 ddh0fix KLD percentile output (#15999)
2025-09-14 Sigbjørn Skjæretmodel : add grok-2 support (#15539)
2025-09-14 Sigbjørn Skjæretserver : only attempt to enable thinking if using jinja...
2025-09-14 Georgi Gerganovmetal : remove memory pools (#15966)
2025-09-14 Adamrocm.Dockerfile: added gfx1200,gfx1201 architectures...
2025-09-14 Ruben OrtlamVulkan: Clean up mul_mm shader (#15987)
2025-09-14 lcybuild: fix the build failures of Windows HIP release...
2025-09-14 Georgi Gerganovmetal : fix kernel requirements (#15983)
2025-09-14 Radoslav Gerganovrpc : fix regression when --device is used (#15981)
2025-09-14 Diego Devesareleases : update ROCM, add gfx1200, gfx1201, gfx1151...
2025-09-14 Radoslav Gerganovdoc : update documentation for --tensor-split (#15980)
2025-09-14 Aaron Teoggml-zdnn: rm user mapped buffers (#15965)
2025-09-13 Jeff Bolzvulkan: fix failing dequant shaders (#15862)
2025-09-13 Jeff Bolzvulkan: initialize vulkan-hpp to allow using extension...
2025-09-13 Diego Devesallama : allow using iGPUs with --device (#15951)
2025-09-13 Georgi Gerganovmetal : refactor kernel loading (#15964)
2025-09-13 Georgi Gerganovmetal : allow ops to run concurrently (#15929)
2025-09-13 Georgi Gerganovmetal : fix memory leaks (#15962)
2025-09-12 Aaron Teoggml-zdnn: fix #15414, activate FP16 and BF16 accelerat...
2025-09-12 Eric CurtinAdd docker protocol support for llama-server model...
2025-09-12 Haiyue Wangcontext : remove redundant explicit casting to the...
2025-09-12 Georgi Gerganovserver : adjust prompt similarity thold + add logs...
2025-09-12 Ruben OrtlamVulkan iGPU device selection overhaul and PCI ID API...
2025-09-12 Mathieu Baudiervulkan: Make device memory check more portable (#15939)
next