Historique des commits

Auteur SHA1 Message Date
  Bruce MacDonald 5d71bda478 close llm on interrupt (#577) il y a 1 an
  Michael Yang 82f5b66c01 register HEAD /api/tags il y a 1 an
  Michael Yang c986694367 fix HEAD / request il y a 1 an
  Bruce MacDonald 4cba75efc5 remove tmp directories created by previous servers (#559) il y a 1 an
  Michael Yang 1fabba474b refactor default allow origins il y a 1 an
  Bruce MacDonald 1255bc9b45 only package 11.8 runner il y a 1 an
  Patrick Devine 80dd44e80a Cmd changes (#541) il y a 1 an
  Bruce MacDonald f221637053 first pass at linux gpu support (#454) il y a 1 an
  Patrick Devine e7e91cd71c add autoprune to remove unused layers (#491) il y a 1 an
  Patrick Devine 790d24eb7b add show command (#474) il y a 1 an
  Michael Yang 681f3c4c42 fix num_keep il y a 1 an
  Michael Yang eeb40a672c fix list models for windows il y a 1 an
  Michael Yang 0f541a0367 s/ListResponseModel/ModelResponse/ il y a 1 an
  Bruce MacDonald 42998d797d subprocess llama.cpp server (#401) il y a 1 an
  Patrick Devine 8bbff2df98 add model IDs (#439) il y a 1 an
  Michael Yang 95187d7e1e build release mode il y a 1 an
  Jeffrey Morgan a9f6c56652 fix `FROM` instruction erroring when referring to a file il y a 1 an
  Ryan Baker 0a892419ad Strip protocol from model path (#377) il y a 1 an
  Bruce MacDonald 326de48930 use loaded llm for embeddings il y a 1 an
  Patrick Devine d9cf18e28d add maximum retries when pushing (#334) il y a 1 an
  Michael Yang 6517bcc53c Merge pull request #290 from jmorganca/add-adapter-layers il y a 1 an
  Michael Yang 6a6828bddf Merge pull request #167 from jmorganca/decode-ggml il y a 1 an
  Jeffrey Morgan 040a5b9750 clean up cli flags il y a 1 an
  Michael Yang 6de5d032e1 implement loading ggml lora adapters through the modelfile il y a 1 an
  Michael Yang fccf8d179f partial decode ggml bin for more info il y a 1 an
  Bruce MacDonald 4b3507f036 embeddings endpoint il y a 1 an
  Bruce MacDonald 868e3b31c7 allow for concurrent pulls of the same files il y a 1 an
  Bruce MacDonald 09d8bf6730 fix build errors il y a 1 an
  Bruce MacDonald 7a5f3616fd embed text document in modelfile il y a 1 an
  Jeffrey Morgan cff002b824 use content type `application/x-ndjson` for streaming responses il y a 1 an