]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2024-11-29 Alberto Cabrera... sycl : Reroute permuted mul_mats through oneMKL (#10408)
2024-11-29 Chenguang LiCANN: RoPE operator optimization (#10563)
2024-11-29 Jeff Bolzvulkan: get the first command buffer submitted sooner...
2024-11-29 Ting Loullava: return false instead of exit (#10546)
2024-11-28 Georgi Gerganovggml : remove redundant copyright notice + update authors
2024-11-28 Georgi Gerganovllama : add missing model types
2024-11-28 Xuan Son Nguyenserver : (tests) don't use thread for capturing stdout...
2024-11-28 Johannes Gäßlercommon: fix warning message when no GPU found (#10564)
2024-11-28 Random Flydocs: fix outdated usage of llama-simple (#10565)
2024-11-28 Diego Devesaci : fix tag name in cuda and hip releases (#10566)
2024-11-28 Georgi Gerganovggml : fix row condition for i8mm kernels (#10561)
2024-11-28 Georgi Gerganovcmake : fix ARM feature detection (#10543)
2024-11-28 Shupei Fanggml-cpu: support IQ4_NL_4_4 by runtime repack (#10541)
2024-11-28 Sergio Lópezkompute : improve backend to pass test_backend_ops...
2024-11-28 Ruixin HuangCANN: Update cann.md to display correctly in CLion...
2024-11-28 leo-ponyCANN: Fix SOC_TYPE compile bug (#10519)
2024-11-28 Chenguang LiCANN: ROPE operator optimization (#10540)
2024-11-27 Xuan Son Nguyencommon : fix duplicated file name with hf_repo and...
2024-11-27 uvosAdd some minimal optimizations for CDNA (#10498)
2024-11-27 Diego Devesaci : faster CUDA toolkit installation method and use...
2024-11-27 Georgi Gerganovmetal : fix group_norm support condition (#0)
2024-11-27 Georgi Gerganovsync : ggml
2024-11-27 Frankie RobertsonDo not include arm_neon.h when compiling CUDA code...
2024-11-27 Jeff Bolzvulkan: define all quant data structures in types.comp...
2024-11-27 Jeff Bolzvulkan: Handle GPUs with less shared memory (#10468)
2024-11-27 Jeff Bolzvulkan: further optimize q5_k mul_mat_vec (#10479)
2024-11-27 Jeff Bolzvulkan: skip integer div/mod in get_offsets for batch_i...
2024-11-27 Jeff Bolzvulkan: optimize Q2_K and Q3_K mul_mat_vec (#10459)
2024-11-26 Diego Devesaci : fix cuda releases (#10532)
2024-11-26 Shane AAdd OLMo 2 model in docs (#10530)
2024-11-26 Diego Devesaci : remove nix workflows (#10526)
2024-11-26 Diego Devesallama : disable warnings for 3rd party sha1 dependency...
2024-11-26 Tristan DruyenFix HIP flag inconsistency & build docs (#10524)
2024-11-26 R0CKSTARmtgpu: Add MUSA_DOCKER_ARCH in Dockerfiles && update...
2024-11-26 Jeff Bolzvulkan: fix group_norm (#10496)
2024-11-26 Xuan Son Nguyenserver : replace behave with pytest (#10416)
2024-11-26 Neo Zhang Jianyurestore the condistion to build & update pacakge when...
2024-11-26 Georgi Gerganovcmake : enable warnings in llama (#10474)
2024-11-26 Diego Devesaci : publish the docker images created during scheduled...
2024-11-26 Diego Devesaci : add ubuntu cuda build, build with one arch on...
2024-11-26 Charles Xuggml-cpu: cmake add arm64 cpu feature check for macos...
2024-11-26 Georgi Gerganovserver : fix parallel speculative decoding (#10513)
2024-11-26 Georgi Gerganovspeculative : simplify the implementation (#10504)
2024-11-26 Shanshan ShenCANN: Improve the Inferencing Performance for Ascend...
2024-11-26 Chenguang LiCANN: RoPE and CANCAT operator optimization (#10488)
2024-11-26 Junil Kimvulkan: Fix a vulkan-shaders-gen arugment parsing error...
2024-11-25 Eric CurtinIntroduce llama-run (#10291)
2024-11-25 Diego Devesaci : build docker images only once daily (#10503)
2024-11-25 Georgi Gerganovserver : add more information about error (#10455)
2024-11-25 Georgi Gerganovserver : enable cache_prompt by default (#10501)
2024-11-25 Georgi Gerganovmetal : enable mat-vec kernels for bs <= 4 (#10491)
2024-11-25 Shane ARename Olmo1124 to Olmo2 (#10500)
2024-11-25 Diego Devesallama : accept a list of devices to use to offload...
2024-11-25 Johannes GäßlerGithub: update issue templates [no ci] (#10489)
2024-11-25 bruceproAdd download chat feature to server chat (#10481)
2024-11-25 Georgi Gerganovserver : add speculative decoding support (#10455)
2024-11-25 Diego Devesaggml : add support for dynamic loading of backends...
2024-11-25 Georgi Gerganovtests : fix compile warning
2024-11-25 Georgi Gerganovmetal : minor code formatting
2024-11-25 Neo Zhang Jianyu[SYCL] Fix building Win package for oneAPI 2025.0 updat...
2024-11-25 Georgi Gerganovspeculative : refactor and add a simpler example (...
2024-11-24 Georgi Gerganovflake.lock: Update (#10470)
2024-11-24 Diego Devesallama : fix op mul check with command-r-plus (#10476)
2024-11-24 Gabe Goodhartconvert : XLMRoberta Type Vocab Size (#10458)
2024-11-24 momongafix gguf-py: Conversion error when multiple licenses...
2024-11-23 Diego Devesaggml : do not use ARM features not included in the...
2024-11-22 蕭澧邦ci: Update oneAPI runtime dll packaging (#10428)
2024-11-22 Johannes GäßlerGitHub: ask for more info in issue templates (#10426)
2024-11-22 leo-ponyCANN: Support Ascend310P to accelerate F32 and F16...
2024-11-21 Diego Devesacuda : optimize argmax (#10441)
2024-11-21 Georgi Gerganovllama : handle KV shift for recurrent models (#10402)
2024-11-21 Georgi Gerganovsync : ggml
2024-11-21 slarenggml/sched : do not skip views in pre-assignments
2024-11-21 Johannes Gäßlerggml-opt: fix data corruption (ggml/1022)
2024-11-20 Jeff Bolzvulkan: predicate max operation in soft_max shaders...
2024-11-20 bandoticmake: add link dependencies to cmake find pkg (#10433)
2024-11-20 Diego Devesallama : add .clang-format file (#10415)
2024-11-20 Jeff Bolzvulkan: copy iq4_nl LUT into shared memory (#10409)
2024-11-20 Jeff Bolzvulkan: further optimize mul_mat_vec using larger loads...
2024-11-20 Neo Zhang Jianyuupdate rel to 4040 (#10395)
2024-11-19 Anthony Van... Fix missing file renames in Makefile due to changes...
2024-11-19 haopengadd cmake rvv support (#10411)
2024-11-19 Georgi Gerganovsync : ggml
2024-11-19 Plamen Minevmetal : fox offset integer overflows in im2col (ggml...
2024-11-19 PABmetal : add `GGML_UNARY_OP_ELU` kernel (ggml/1018)
2024-11-19 蕭澧邦cmake: force MSVC compiler charset to utf-8 (#9989)
2024-11-19 bandotiAdd required ggml-base and backend libs to cmake pkg...
2024-11-19 Diego Devesacuda : fix CUDA_FLAGS not being applied (#10403)
2024-11-19 Georgi Gerganovllama : add check for KV cache shifts (#10401)
2024-11-19 Shane Allama : add OLMo November 2024 support (#10394)
2024-11-19 Romain Biessysycl : Add option to set the SYCL architecture for...
2024-11-19 Jeff Bolzvulkan: Optimize soft_max (#10301)
2024-11-19 Alberto Cabrera... sycl: Revert MUL_MAT_OP support changes (#10385)
2024-11-18 Diego Devesacuda : only use native when supported by cmake (#10389)
2024-11-18 bandotiSkip searching root path for cross-compile builds ...
2024-11-18 Jeff Bolzvulkan: remove use of null initializer (#10372)
2024-11-18 Georgi Gerganovflake.lock: Update (#10346)
2024-11-18 0cc4mVulkan: Fix device info output format specifiers (...
2024-11-17 Johannes Gäßlerdocker: use GGML_NATIVE=OFF (#10368)
2024-11-17 Johannes GäßlerCUDA: fix MMV kernel being used for FP16 src1 (#10357)
next