Code

[pull] master from ggerganov:master#151

Code

Closed

pull[bot] wants to merge 46 commits intoteleprint-me:masterteleprint-me/llama.cpp:masterfrom ggerganov:masterggerganov/llama.cpp:masterCopy head branch name to clipboard

+5,959-8,191

Commits

Commits on Nov 15, 2024

• backend cpu: add online flow for aarch64 Q4_0 GEMV/GEMM kernels (#9921)

  Show description for 1607a5e
  

  chaxu01

  and

  slaren

  authored
  1607a5e

• sycl: Use syclcompat::dp4a (#10267)

  Show description for 5a54af4
  Rbiessy

  authored
  5a54af4

• scripts : fix regex in sync [no ci]

  ggerganov

  committed
  4802ad3

• cann: dockerfile and doc adjustment (#10302)

  Show description for 231f936
  

  noemotiovon

  and

  noemotiovon

  authored
  231f936

• server : (web UI) add copy button for code block, fix api key (#10242)

  Show description for 9901068
  ngxson

  authored
  9901068

• sycl: Update Intel docker images to use DPC++ 2025.0 (#10305)

  Rbiessy

  authored
  57f8355

• ci: build test musa with cmake (#10298)

  Show description for f0204a0
  yeahdongcn

  authored
  f0204a0

• AVX BF16 and single scale quant optimizations (#10212)

  Show description for 1842922
  netrunnereve

  authored
  1842922

• sync : ggml

  ggerganov

  committed
  cbf5541

• ggml : vulkan logs (whisper/2547)

  

  thewh1teagle

  authored and

  ggerganov

  committed
  3225008

• cmake : fix ppc64 check (whisper/0)

  Show description for 09ecbcb
  ggerganov

  committed
  09ecbcb

• ggml : fix some build issues

  

  slaren

  authored and

  ggerganov

  committed
  883d206

• scripts: update compare-llama-bench.py (#10319)

  JohannesGaessler

  authored
  4047be7

• Make updates to fix issues with clang-cl builds while using AVX512 flags (#10314)

  Srihari-mcw

  authored
  74d73dc

Commits on Nov 16, 2024

• llama : save number of parameters and the size in llama_model (#10286)

  Show description for 89e4caa
  FirstTimeEZ

  authored
  89e4caa

• ggml : optimize Q4_0 into Q4_0_X_Y repack (#10324)

  eddnjjn

  authored
  1e58ee1

• vulkan : add cmake preset debug/release (#10306)

  FirstTimeEZ

  authored
  dd3a6ce

• vulkan: Optimize some mat-vec mul quant shaders (#10296)

  Show description for 772703c
  jeffbolznv

  authored
  772703c

• scripts : fix missing key in compare-llama-bench.py (#10332)

  ggerganov

  authored
  f245cc2

• server: (web UI) Add samplers sequence customization (#10255)

  Show description for bcdb7a2
  

  MaggotHATE

  and

  ngxson

  authored
  bcdb7a2

• make : auto-determine dependencies (#0)

  ggerganov

  committed
  8ee0d09

• llamafile : fix include path (#0)

  Show description for db4cfd5
  ggerganov

  committed
  db4cfd5

• llama/ex: remove --logdir argument (#10339)

  JohannesGaessler

  authored
  4e54be0

• docs : vulkan build instructions to use git bash mingw64 (#10303)

  FirstTimeEZ

  authored
  0fff7fd

Commits on Nov 17, 2024

• scripts : update sync

  ggerganov

  committed
  5c9a8b2

• ggml: new optimization interface (ggml/988)

  

  JohannesGaessler

  authored and

  ggerganov

  committed
  8a43e94

• ggml : fix compile warnings (#0)

  Show description for 68fcb47
  ggerganov

  committed
  68fcb47

• tests : remove test-grad0

  ggerganov

  committed
  84274a1

• make : add ggml-opt (#0)

  Show description for a4200ca
  ggerganov

  committed
  a4200ca

• ggml : adapt AMX to tensor->grad removal (#0)

  Show description for 5d9e599
  ggerganov

  committed
  5d9e599

• ggml : inttypes.h -> cinttypes (#0)

  Show description for 24203e9
  ggerganov

  committed
  24203e9

• ggml : fix possible buffer use after free in sched reserve (#9930)

  slaren

  authored
  eda7e1d

• CMake: default to -arch=native for CUDA build (#10320)

  JohannesGaessler

  authored
  467576b

• CUDA: remove DMMV, consolidate F16 mult mat vec (#10318)

  JohannesGaessler

  authored
  c3ea58a

• ggml : fix undefined reference to 'getcpu' (#10354)

  Show description for a431782
  FirstTimeEZ

  authored
  a431782

• metal : refactor kernel args into structs (#10238)

  Show description for cf32a9b
  ggerganov

  authored
  cf32a9b

• gitignore : ignore local run scripts [no ci]

  ggerganov

  committed
  20a780c

• llama : only use default buffer types for the KV cache (#10358)

  slaren

  authored
  be5cacc

• CMake: fix typo in comment [no ci] (#10360)

  JohannesGaessler

  authored
  ce2e59b

• CUDA: fix MMV kernel being used for FP16 src1 (#10357)

  JohannesGaessler

  authored
  76e9e58

• docker: use GGML_NATIVE=OFF (#10368)

  JohannesGaessler

  authored
  75207b3

Commits on Nov 18, 2024

• Vulkan: Fix device info output format specifiers (#10366)

  Show description for 9b75f03
  0cc4m

  authored
  9b75f03

• flake.lock: Update (#10346)

  Show description for 2eb76b2
  

  ggerganov

  and

  github-actions[bot]

  authored
  2eb76b2

• vulkan: remove use of null initializer (#10372)

  Show description for f139d2e
  jeffbolznv

  authored
  f139d2e

• Skip searching root path for cross-compile builds (#10383)

  bandoti

  authored
  531cb1c

• cuda : only use native when supported by cmake (#10389)

  slaren

  authored
  d3481e6