Commits

Breadcrumbs

History for

on2e66683

User selector

All users

Datepicker

All time

Commit History

Commits on Jun 1, 2024

• server : new UI (ggml-org#7633)

  Show description for 2e66683
  mounta11n

  authored
  2e66683

  Copy full SHA for 2e66683

• CUDA: quantized KV support for FA vec (ggml-org#7527)

  Show description for 9b59641
  JohannesGaessler

  authored
  9b59641

  Copy full SHA for 9b59641

Commits on May 31, 2024

• Improve HIP compatibility (ggml-org#7672)

  daniandtheweb

  authored
  30e238b

  Copy full SHA for 30e238b

Commits on May 27, 2024

• make: add --device-debug to NVCC debug flags (ggml-org#7542)

  JohannesGaessler

  authored
  10b1e45

  Copy full SHA for 10b1e45

Commits on May 23, 2024

• ggml : drop support for QK_K=64 (ggml-org#7473)

  Show description for e84b71c
  ggerganov

  authored
  e84b71c

  Copy full SHA for e84b71c

Commits on May 20, 2024

• ggml : add loongarch lsx and lasx support (ggml-org#6454)

  Show description for 65c5820
  

  junchao-loongson

  and

  MQ-mengqing

  authored
  65c5820

  Copy full SHA for 65c5820

Commits on May 19, 2024

• llama : remove MPI backend (ggml-org#7395)

  slaren

  authored
  d359f30

  Copy full SHA for d359f30

Commits on May 17, 2024

• ROCm: use native CMake HIP support (ggml-org#5966)

  Show description for 82ca83d
  GZGavinZhao

  authored
  82ca83d

  Copy full SHA for 82ca83d

Commits on May 8, 2024

• Introduction of CUDA Graphs to LLama.cpp (ggml-org#6766)

  Show description for bc4bba3
  

  agray3

  and

  slaren

  authored
  bc4bba3

  Copy full SHA for bc4bba3

Commits on May 4, 2024

• tests : add test-tokenizer-0.sh + fix some tokenizers (ggml-org#7036)

  Show description for 92139b9
  ggerganov

  authored
  92139b9

  Copy full SHA for 92139b9

Commits on Apr 29, 2024

• llama : fix BPE pre-tokenization (ggml-org#6920)

  Show description for f4ab2a4
  

  3 peopleauthored
  f4ab2a4

  Copy full SHA for f4ab2a4

• make : change GNU make default CXX from g++ to c++ (ggml-org#6966)

  przemoc

  authored
  577277f

  Copy full SHA for 577277f

Commits on Apr 26, 2024

• quantize: add imatrix and dataset metadata in GGUF (ggml-org#6658)

  Show description for 0c4d489
  

  3 peopleauthored
  0c4d489

  Copy full SHA for 0c4d489

Commits on Apr 22, 2024

• llamafile : improve sgemm.cpp (ggml-org#6796)

  Show description for 192090b
  jart

  authored
  192090b

  Copy full SHA for 192090b

Commits on Apr 21, 2024

• build: generate hex dump of server assets during build (ggml-org#6661)

  Show description for 5cf5e7d
  ochafik

  authored
  5cf5e7d

  Copy full SHA for 5cf5e7d

• llama : add option to render special/control tokens (ggml-org#6807)

  Show description for 40f74e4
  ggerganov

  authored
  40f74e4

  Copy full SHA for 40f74e4

Commits on Apr 17, 2024

• llamafile : tmp disable + build sgemm.o when needed (ggml-org#6716)

  Show description for 3b8f1ec
  ggerganov

  authored
  3b8f1ec

  Copy full SHA for 3b8f1ec

Commits on Apr 16, 2024

• ggml : fix llamafile sgemm wdata offsets (ggml-org#6710)

  Show description for 666867b
  ggerganov

  authored
  666867b

  Copy full SHA for 666867b

• ggml : add llamafile sgemm (ggml-org#6414)

  Show description for 8cc91dc
  jart

  authored
  8cc91dc

  Copy full SHA for 8cc91dc

Commits on Apr 15, 2024

• main: add --json-schema / -j flag (ggml-org#6659)

  Show description for 7593639
  ochafik

  authored
  7593639

  Copy full SHA for 7593639

Commits on Apr 11, 2024

• Refactor Error Handling for CUDA (ggml-org#6575)

  Show description for a474f50
  

  nneubacher

  and

  JohannesGaessler

  authored
  a474f50

  Copy full SHA for a474f50

• eval-callback: Example how to use eval callback for debugging (ggml-org#6576)

  Show description for b804b1e
  

  3 peopleauthored
  b804b1e

  Copy full SHA for b804b1e

Commits on Apr 6, 2024

• Tests: Added integration tests for GBNF parser (ggml-org#6472)

  Show description for 57dd02c
  HanClinto

  authored
  57dd02c

  Copy full SHA for 57dd02c

Commits on Apr 4, 2024

• examples : add GBNF validator program (ggml-org#5948)

  Show description for 9b84ae1
  HanClinto

  authored
  9b84ae1

  Copy full SHA for 9b84ae1

Commits on Mar 27, 2024

• make : whitespace

  ggerganov

  committed
  3a03459

  Copy full SHA for 3a03459

Commits on Mar 26, 2024

• wpm : portable unicode tolower (ggml-org#6305)

  Show description for 32c8486
  cebtenzzre

  authored
  32c8486

  Copy full SHA for 32c8486

• cuda : rename build flag to LLAMA_CUDA (ggml-org#6299)

  slaren

  authored
  2803459

  Copy full SHA for 2803459

Commits on Mar 25, 2024

• cuda : refactor into multiple files (ggml-org#6269)

  slaren

  authored
  ae1f211

  Copy full SHA for ae1f211

• examples : add "retrieval" (ggml-org#6193)

  Show description for 64e7b47
  

  mscheong01

  and

  ggerganov

  authored
  64e7b47

  Copy full SHA for 64e7b47

Commits on Mar 23, 2024

• split: add gguf-split in the make build target (ggml-org#6262)

  phymbert

  authored
  21cad01

  Copy full SHA for 21cad01

• lookup: complement data from context with general text statistics (ggml-org#5479)

  Show description for 50ccaf5
  JohannesGaessler

  authored
  50ccaf5

  Copy full SHA for 50ccaf5

Commits on Mar 22, 2024

• cuda : add LLAMA_CUDA_NO_PEER_COPY to workaround broken ROCm p2p copy (ggml-org#6208)

  Show description for 2f0e81e
  slaren

  authored
  2f0e81e

  Copy full SHA for 2f0e81e

Commits on Mar 21, 2024

• json-schema-to-grammar improvements (+ added to server) (ggml-org#5978)

  Show description for 5b7b0ac
  ochafik

  authored
  5b7b0ac

  Copy full SHA for 5b7b0ac

Commits on Mar 19, 2024

• gguf-split: split and merge gguf per batch of tensors (ggml-org#6135)

  Show description for d0d5de4
  

  phymbert

  and

  ggerganov

  authored
  d0d5de4

  Copy full SHA for d0d5de4

Commits on Mar 17, 2024

• common: llama_load_model_from_url using --model-url (ggml-org#6098)

  Show description for d01b3c4
  

  phymbert

  and

  ggerganov

  authored
  d01b3c4

  Copy full SHA for d01b3c4

Pagination

Previous

Next