Code

merge upstream#41

Code

Merged

l3utterfly merged 13 commits intolayla-buildl3utterfly/llama.cpp:layla-buildfrom masterl3utterfly/llama.cpp:masterCopy head branch name to clipboardOct 3, 2024

+4,493-4,715

Commits

Commits on Oct 1, 2024

• metal : reduce command encoding overhead (ggerganov#9698)

  Show description for cad341d
  ggerganov

  authored
  cad341d

• ggml: fix gradient allocation logic (ggml/966)

  Show description for 7254cdf
  

  JohannesGaessler

  authored and

  ggerganov

  committed
  7254cdf

• ggml : fix ggml_cast (ggml/973)

  

  iboB

  authored and

  ggerganov

  committed
  6c53224

• vulkan : mul_mat: fix UB with small warps (ggml/952)

  Show description for cb00020
  

  smeso

  authored and

  ggerganov

  committed
  cb00020

• test: fix OPT_STEP_ADAMW for test-backend-ops (ggml/974)

  

  JohannesGaessler

  authored and

  ggerganov

  committed
  e98c1c1

• sync : ggml

  ggerganov

  committed
  f1b8c42

• Update README.md (ggerganov#9591)

  Show description for 3f1ae2e
  32bitmicro

  authored
  3f1ae2e

Commits on Oct 2, 2024

• examples : remove benchmark (ggerganov#9704)

  Show description for 148844f
  ggerganov

  authored
  148844f

• gguf-split : improve --split and --merge logic (ggerganov#9619)

  Show description for 76b37d1
  

  3 peopleauthored
  76b37d1

• vulkan : do not use tensor->extra (ggerganov#9407)

  Show description for 00b7317
  

  rgerganov

  and

  0cc4m

  authored
  00b7317

• [SYCL] Initial cmake support of SYCL for AMD GPUs (ggerganov#9658)

  Show description for f536f4c
  Alcpz

  authored
  f536f4c

• llama : reduce compile time and binary size (ggerganov#9712)

  Show description for a39ab21
  ngxson

  authored
  a39ab21

• ggml-backend : add device and backend reg interfaces (ggerganov#9707)

  Show description for c83ad6d
  

  slaren

  and

  JohannesGaessler

  authored
  c83ad6d