Commits

Branch selector

 59f4db1

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 4, 2024

• ggml : add predefined list of CPU backend variants to build (ggerganov#10626)

  Show description for 59f4db1
  slaren

  authored
  59f4db1

• ggml-cpu : fix HWCAP2_I8MM value (ggerganov#10646)

  slaren

  authored
  2803540

• Fix HF repo commit to clone lora test models (ggerganov#10649)

  ltoniazzi

  authored
  253b7fd

• llama: Support MiniCPM-1B (with & w/o longrope) (ggerganov#10559)

  JFLFY2255

  authored
  8d0cfd5

• vulkan: Implement "fast divide" (mul+shift) for unary ops like copy (ggerganov#10642)

  jeffbolznv

  authored
  2759916

• SYCL : Move to compile time oneMKL interface backend selection for NVIDIA backend (ggerganov#10584)

  Show description for 40c6d79
  s-Nick

  authored
  40c6d79

• fix typo of README.md (ggerganov#10605)

  WrRan

  authored
  98036d5

• Avoid using __fp16 on ARM with old nvcc (ggerganov#10616)

  frankier

  authored
  cd2f37b

• Add docs for creating a static build (ggerganov#10268) (ggerganov#10630)

  Show description for da6aac9
  

  mostlygeek

  and

  slaren

  authored
  da6aac9

• clip : add sycl support (ggerganov#10574)

  Show description for 01e6d9b
  

  piDack

  and

  piDack

  authored
  01e6d9b

Commits on Dec 3, 2024

• vulkan: optimize and reenable split_k (ggerganov#10637)

  Show description for cc98896
  jeffbolznv

  authored
  cc98896

• server : (web ui) Various improvements, now use vite as bundler (ggerganov#10599)

  Show description for 91c36c2
  ngxson

  authored
  91c36c2

• scripts : remove amx sync

  Show description for 1cd3df4
  ggerganov

  committed
  1cd3df4

• sync : ggml

  ggerganov

  committed
  c505471

• CUDA: remove unnecessary warp reduce in FA (ggml/1032)

  Show description for e9e661b
  

  2 peopleauthored and

  ggerganov

  committed
  e9e661b

• feat: add GGML_UNARY_OP_ARGMAX Metal kernel (ggml/1019)

  Show description for efb6ae9
  

  2 peopleauthored and

  ggerganov

  committed
  efb6ae9

• metal : add GGML_OP_CONV_TRANSPOSE_1D kernels (ggml/1026)

  Show description for 667d70d
  

  PABannier

  authored and

  ggerganov

  committed
  667d70d

• llama : add missing LLAMA_API for llama_chat_builtin_templates (ggerganov#10636)

  ngxson

  authored
  3b4f2e3

• readme : add option, update default value, fix formatting (ggerganov#10271)

  Show description for 82bca22
  pothitos

  authored
  82bca22

• metal : small-batch mat-mul kernels (ggerganov#10581)

  Show description for 0115df2
  ggerganov

  authored
  0115df2

• github : minify link [no ci] (revert)

  Show description for 515d4e5
  ggerganov

  committed
  515d4e5

• github : minify link [no ci]

  ggerganov

  committed
  844e2e1

• server : fix default draft model parameters (ggerganov#10586)

  Show description for 70b98fa
  ggerganov

  authored
  70b98fa

Commits on Dec 2, 2024

• llama : add enum for built-in chat templates (ggerganov#10623)

  Show description for 642330a
  ngxson

  authored
  642330a

• make : deprecate (ggerganov#10514)

  Show description for 8648c52
  

  ggerganov

  and

  slaren

  authored
  8648c52

• server: Add "tokens per second" information in the backend (ggerganov#10548)

  Show description for 64ed209
  

  lhpqaq

  and

  ngxson

  authored
  64ed209

• SYCL: Fix and switch to GGML_LOG system instead of fprintf (ggerganov#10579)

  Show description for 991f8aa
  qnixsynapse

  authored
  991f8aa

• contrib : refresh (ggerganov#10593)

  Show description for 4cb003d
  ggerganov

  authored
  4cb003d

Commits on Dec 1, 2024

• Add mistral-v1, mistral-v3, mistral-v3-tekken and mistral-v7 chat template types (ggerganov#10572)

  Show description for 917786f
  jukofyork

  authored
  917786f

• grammars : add English-only grammar (ggerganov#10612)

  ggerganov

  authored
  5e1ed95

• ci: add error handling for Python venv creation in run.sh (ggerganov#10608)

  wangqin0

  authored
  5c7a5aa

• ggml : automatic selection of best CPU backend (ggerganov#10606)

  Show description for 3420909
  slaren

  authored
  3420909

• server : bind to any port when specified (ggerganov#10590)

  alek3y

  authored
  86dc11c

• readme : update the usage section with examples (ggerganov#10596)

  Show description for 6acce39
  ggerganov

  authored
  6acce39

• build: update Makefile comments for C++ version change (ggerganov#10598)

  wangqin0

  authored
  43957ef

Pagination

Previous

Next