]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2025-11-14 Aleksander... Better UX for handling multiple attachments in WebUI...
2025-11-13 Alberto Cabrera... ggml-cpu: handle 3d tensors in repack mat_mul (#17241)
2025-11-13 Xuan-Son Nguyenserver: fixing naming conflict res_error (#17243)
2025-11-13 Piotr Wilkin... ggml : add ops SOFTPLUS, EXPM1, TRI, SOLVE_TRI, CUMSUM...
2025-11-13 Ruben Ortlamvulkan: remove shell call from vulkan-shaders-gen tool...
2025-11-13 Diego Devesasched : fix reserve ignoring user tensor assignments...
2025-11-13 ixgbeggml-cpu : add RISC-V vector intrinsic support for...
2025-11-13 bagheerametal: accelerated conv2d (#17175)
2025-11-13 Georgi GerganovRevert "ggml-cpu: handle 3d tensors in repack mat_mul...
2025-11-13 Diego Devesaggml-cpu : use template for argsort (#17222)
2025-11-13 TecJeshCANN: Add cross_entropy_loss op support (#16886)
2025-11-13 Aman GuptaCUDA: fuse rope + set_rows (#16884)
2025-11-13 Neo Zhang Jianyuupdate SYCL support OPs (#17208)
2025-11-12 o7sivocab : correct bounds check for UGM XCDA array access...
2025-11-12 Johannes GäßlerCUDA: static assert to prevent misuse of memcpy_1 ...
2025-11-12 Mike Abbottdocker : preserve .so symlinks for docker container...
2025-11-12 Georgi Gerganovggml : use std::sort in ggml_argsort CPU implementation...
2025-11-12 Aleksander... Update packages + upgrade Storybook to v10 (#17201)
2025-11-12 Xuan-Son Nguyenserver: (refactor) implement generator-based API for...
2025-11-12 Xuan-Son Nguyenci: add check vendor job (#17179)
2025-11-12 Xuan-Son Nguyenserver: move res_error/res_ok to static function (...
2025-11-12 Alberto Cabrera... ggml-cpu: handle 3d tensors in repack mat_mul (#17030)
2025-11-12 Adrien Gallouëtcmake : cleanup (#17199)
2025-11-12 Adrien Gallouëtcmake : move OpenSSL linking to vendor/cpp-httplib...
2025-11-12 TecJeshCANN: Add L2_NORM op support (#16856)
2025-11-12 Neo Zhang Jianyu[SYCL]fix ci crash about SSM_CONV (#17169)
2025-11-12 Raul TorresCANN: GGML_CANN_ACL_GRAPH works only USE_ACL_GRAPH...
2025-11-11 Max Krasnyanskyhexagon: various Op fixes (#17135)
2025-11-11 Evedisable rms norm mul rope for chips with no fp16 rte...
2025-11-11 sudhiarmci: add Arm-hosted Graviton4 runner (#17021)
2025-11-11 Xuan-Son Nguyenvendor: split httplib to cpp/h files (#17150)
2025-11-11 ixgbeggml-cpu : add RISC-V RVV (Zvfh) optimization for FP16...
2025-11-11 dudutaggml-cpu: templateify ggml_compute_forward_rope_f32...
2025-11-11 Charles Xukleidiai: add optimized per-channel kernels for Q8_0...
2025-11-11 Mike Abbottcmake : add version to all shared object files (#17091)
2025-11-11 Nicolas B.... Install rpc-server when GGML_RPC is ON. (#17149)
2025-11-11 levkroppconvert : register UMT5Model architecture for T5 conver...
2025-11-10 lhezopencl: add fastdiv and use it in set_rows, ported...
2025-11-10 Sigbjørn Skjæretmodels : move build_inp_out_ids outside loop (#17151)
2025-11-10 Max Krasnyanskycpu: skip NOPs to avoid barriers (#17133)
2025-11-10 Georgi Gerganovmetal : cap threadgroups size of set_rows (#17146)
2025-11-10 Adrien Gallouëtggml-cpu : inspect -march and -mcpu to found the CPU...
2025-11-10 Ruben Ortlamvulkan: check glslc executable string (#17144)
2025-11-10 Ruben Ortlamvulkan: fix validation issue introduced by #16868 ...
2025-11-10 Gabe Goodhartmemory: Hybrid context shift (#17009) upstream/0.0.7011
2025-11-10 Georgi Gerganovmetal : enable tensor API for A19 (#17087)
2025-11-10 fj-y-saitoarm64: add i8mm route with SVE ggml_vec_dot_q4_K_q8_K...
2025-11-10 Georgi Gerganovbatched-bench : add "separate text gen" mode (#17103)
2025-11-10 Xuan-Son Nguyenmtmd: fix patch_size initialized to random value in...
2025-11-10 Georgi Gerganoveditorconfig : ignore benches/ (#17140)
2025-11-10 Aclycuda/vulkan : bicubic interpolation (#17022)
2025-11-10 Georgi Gerganovbenches : add eval results (#17139)
2025-11-09 Georgi Gerganovmtmd : fix embedding size for image input (#17123)
2025-11-09 Ruben Ortlamvulkan: fix memory allocations (#17122)
2025-11-09 compiladeconvert : parse safetensors directly (#15667)
2025-11-09 compiladeconvert : handle compressed-tensors quant method (...
2025-11-09 Georgi Gerganovserver : handle failures to restore host cache (#17078)
2025-11-09 Georgi Gerganovbenches : add folder with benchmarks (#16931)
2025-11-09 Eric CurtinSwitch to using Ubuntu 25.10 vulkan/mesa (#16497)
2025-11-09 Ruben Ortlamvulkan: iGPU memory reporting fix (#17110)
2025-11-09 Ruben Ortlamvulkan: fix mmq out of bounds reads (#17108)
2025-11-09 Jeff Bolzvulkan: fuse mul_mat_id + mul (#17095)
2025-11-09 Georgi Gerganovmetal : retain src and dst buffers during async ops...
2025-11-08 Xuan-Son Nguyenarg: add --cache-list argument to list cached models...
2025-11-08 chansikparkwebui: fix keyboard shortcuts for new chat & edit chat...
2025-11-08 Jeff Bolzvulkan: Use spec constants for conv2d s/d/p and kernel...
2025-11-08 Aidanserver: fix correct time_ms calculation in prompt_progr...
2025-11-08 Aman GuptaRevert "CUDA: add expert reduce kernel (#16857)" (...
2025-11-08 Aman GuptaCUDA: skip fusion for repeating adds in bias (#17080)
2025-11-08 SavicStefanvulkan: Increase BK to 32; use BK/4 for non-CM mul_mm...
2025-11-08 Aleksei Nikiforovggml: disable vxe for cross-compilation by default...
2025-11-08 Jeff Bolzvulkan: fuse rms_norm + mul + rope (+ view + set_rows...
2025-11-08 Jeff Bolzvulkan: Fix test-thread-safety crashes (#17024)
2025-11-08 Johannes GäßlerCUDA: fix MMQ stream-k fixup ne1 indices (#17089)
2025-11-08 Reese Levineggml webgpu: faster matrix multiplication/matrix-vector...
2025-11-07 bssrdfCUDA: properly handle nb00=nb02 case for cpy (#17081)
2025-11-07 Aclyvulkan : refactor buffer handling in vk_op_f32 (#16840)
2025-11-07 Johannes GäßlerCUDA: fix should_use_mmvf for ne11 == 1 (#17085)
2025-11-07 Georgi Gerganovbench : cache the llama_context state at computed depth...
2025-11-07 Sigbjørn Skjærethparams : add n_embd_inp() to support extended embed...
2025-11-07 Georgi Gerganovkv-cache : pad the cache size to 256 for performance...
2025-11-07 Adrien GallouëtRevert "ggml-cpu: detect correct cpu flags for arm64...
2025-11-07 ironggml-cpu: detect correct cpu flags for arm64 (#16229...
2025-11-07 Georgi Gerganovserver : print the samplers chain for each request...
2025-11-07 Xuan-Son Nguyencommon: move download functions to download.(cpp|h...
2025-11-06 xctanggml-cpu : optimize RVV q2_k and q3_k kernels (#16887)
2025-11-06 Johannes GäßlerCUDA: fix crash on uneven context without FA (#16988)
2025-11-06 Georgi Gerganovmetal : initial Metal4 tensor API support (#16634)
2025-11-06 Georgi Gerganovserver : disable checkpoints with mtmd (#17045)
2025-11-06 Xuan-Son Nguyenclip: implement minicpm-v sinusoidal embd using GGML...
2025-11-06 YehuditEsycl: add CONCAT operator support (#16047)
2025-11-06 Johannes Gäßlerdocs: explain CUDA 11 compilation [no ci] (#16824)
2025-11-06 l3utterflyggml-hexagon: graceful fallback for older socs where...
2025-11-05 bssrdfimprove CUDA cpy memory bandwidth when copying transpos...
2025-11-05 Jeff Bolzvulkan: Fix GGML_VULKAN_CHECK_RESULTS to better handle...
2025-11-05 Gabe Goodhartexamples(gguf): GGUF example outputs (#17025)
2025-11-05 Xuan-Son Nguyenmtmd: allow QwenVL to process larger image by default...
2025-11-05 Georgi Gerganovserver : do not default to multiple slots with speculat...
2025-11-05 Xuan-Son Nguyenmtmd: improve struct initialization (#16981)
2025-11-05 손희준docs: Clarify the endpoint that webui uses (#17001)
next