Commit History

Автор SHA1 Съобщение Дата
  Georgi Gerganov 9c67c2773d ggml : add Flash Attention (#5021) преди 1 година
  Georgi Gerganov 952d03dbea convert : use utf8 encoding (#7000) преди 1 година
  Olivier Chafik 8843a98c2b Improve usability of --model-url & related flags (#6930) преди 1 година
  Clint Herron b8c1476e44 Extending grammar integration tests (#6644) преди 1 година
  Daniel Bevenius 5539e6fdd1 main : fix typo in comment in main.cpp (#6985) преди 1 година
  Olivier Chafik b8a7a5a90f build(cmake): simplify instructions (`cmake -B build && cmake --build build ...`) (#6964) преди 1 година
  Georgi Gerganov d2c898f746 ci : tmp disable gguf-split (#6983) преди 1 година
  Georgi Gerganov 544f1f10ad ggml : fix __MSC_VER -> _MSC_VER (#6977) преди 1 година
  cpumaxx ffe666572f llava-cli : multiple images (#6969) преди 1 година
  Georgi Gerganov 24affa7db3 readme : update hot topics преди 1 година
  Georgi Gerganov f4ab2a4147 llama : fix BPE pre-tokenization (#6920) преди 1 година
  David Renshaw 3f167476b1 sampling : use std::random_device{}() for default random seed (#6962) преди 1 година
  Christian Zhou-Zheng 3055a41805 convert : fix conversion of some BERT embedding models (#6937) преди 1 година
  Przemysław Pawełczyk 577277ffd2 make : change GNU make default CXX from g++ to c++ (#6966) преди 1 година
  Przemysław Pawełczyk ca7f29f568 ci : add building in MSYS2 environments (Windows) (#6967) преди 1 година
  Johannes Gäßler c4f708a93f llama : fix typo LAMMAFILE -> LLAMAFILE (#6974) преди 1 година
  DAN™ e00b4a8f81 Fix more int overflow during quant (PPL/CUDA). (#6563) преди 1 година
  Xuan Son Nguyen 7bb36ccf91 gguf : enforce that tensor names are unique (#6905) преди 1 година
  Neo Zhang ce023f6f2f add device version in device list (#6959) преди 1 година
  github-actions[bot] 6e472f58e4 flake.lock: Update преди 1 година
  mgroeber9110 4dba7e8114 Replace "alternative" boolean operator in conditional compilation directive (#6949) преди 1 година
  Pierrick Hymbert b7368332e2 ci: server: tests python env on github container ubuntu latest / fix n_predict (#6935) преди 1 година
  agray3 928e0b7013 Reset schedule earlier to allow overlap with ggml graph computation on device (#6933) преди 1 година
  Pierrick Hymbert 0c4d489e29 quantize: add imatrix and dataset metadata in GGUF (#6658) преди 1 година
  slaren 017e6999b5 add basic tensor data validation function (#6884) преди 1 година
  slaren e2764cd7ca gguf : fix mismatch between alloc and free functions (#6929) преди 1 година
  Justine Tunney 4b1c3c98b4 llamafile : use 64-bit integers in sgemm (#6928) преди 1 година
  Pierrick Hymbert bbe3c6e761 ci: server: fix python installation (#6925) преди 1 година
  Pierrick Hymbert 7f5ff558ee server: stop generation at `n_ctx_train` if `n_predict` is not set (#6638) преди 1 година
  Pierrick Hymbert 9e4e077ec5 ci: server: fix python installation (#6922) преди 1 година