Commit History

Author SHA1 Message Date
  jmorganca e86db9381a `avx2` should only add `avx2` 11 months ago
  jmorganca 9fe48978a8 move `runner` package down 11 months ago
  jmorganca 01ccbc07fe replace static build in `llm` 11 months ago
  jmorganca 0110994d06 Initial `llama` Go module 1 year ago
  jmorganca 2ef3a217d1 add sync of llama.cpp 1 year ago
  Michael Yang fccf8d179f partial decode ggml bin for more info 1 year ago
  Bruce MacDonald 984c9c628c fix embeddings invalid values 1 year ago
  Bruce MacDonald 09d8bf6730 fix build errors 1 year ago
  Bruce MacDonald 7a5f3616fd embed text document in modelfile 1 year ago
  Michael Yang f2074ed4c0 Merge pull request #306 from jmorganca/default-keep-system 1 year ago
  Bruce MacDonald a6f6d18f83 embed text document in modelfile 1 year ago
  Jeffrey Morgan 5eb712f962 trim whitespace before checking stop conditions 1 year ago
  Michael Yang 4dc5b117dd automatically set num_keep if num_keep < 0 1 year ago
  Michael Yang b9f4d67554 configurable rope frequency parameters 1 year ago
  Michael Yang c5bcf32823 update llama.cpp 1 year ago
  Michael Yang 74a5f7e698 no gpu for 70B model 1 year ago
  Michael Yang 319f078dd9 remove -Werror 1 year ago
  Jeffrey Morgan 7da249fcc1 only build metal for `darwin,arm` target 1 year ago
  Bruce MacDonald 184ad8f057 allow specifying stop conditions in modelfile 1 year ago
  Michael Yang 3549676678 embed ggml-metal.metal 1 year ago
  Michael Yang fadf75f99d add stop conditions 1 year ago
  Michael Yang ad3a7d0e2c add NumGQA 1 year ago
  Michael Yang cca61181cb sample metrics 1 year ago
  Michael Yang c490416189 lock on llm.lock(); decrease batch size 1 year ago
  Michael Yang f62a882760 add session expiration 1 year ago
  Michael Yang 3003fc03fc update predict code 1 year ago
  Michael Yang 35af37a2cb session id 1 year ago
  Michael Yang 726bc647b2 enable k quants 1 year ago
  Michael Yang cb55fa9270 enable accelerate 1 year ago
  Michael Yang b71c67b6ba allocate a large enough tokens slice 1 year ago