Commit Graph

  • 3bb45f1658 More reasonable defaults Mug 2023-04-10 16:38:45 +02:00
  • 0cccb41a8f Added iterative search to prevent instructions from being echoed, add ignore eos, add no-mmap, fixed 1 character echo too much bug Mug 2023-04-10 16:35:38 +02:00
  • c65a621b6b Add UTF-8 Encoding in read_text. jm12138 2023-04-10 10:28:24 +00:00
  • 241d608bbb Update workflow permissions Andrei Betlen 2023-04-10 02:35:00 -04:00
  • 3d56c3b706 Run tests for pr's to main Andrei Betlen 2023-04-10 02:19:22 -04:00
  • bc02ce353b Bump version v0.1.30 Andrei Betlen 2023-04-10 02:12:19 -04:00
  • 1f67ad2a0b Add use_mmap option Andrei Betlen 2023-04-10 02:11:35 -04:00
  • d41cb0ecf7 Add create release step to workflow Andrei Betlen 2023-04-10 01:54:52 -04:00
  • 8594b8388e Add build and release Andrei Betlen 2023-04-10 01:29:32 -04:00
  • a984f55d79 Quickfix: forgot to clone submodules when building and publishing pypi package v0.1.29 Andrei Betlen 2023-04-10 00:51:25 -04:00
  • 196650ccb2 Update model paths to be more clear they should point to file Andrei Betlen 2023-04-09 22:45:55 -04:00
  • a79d3eb732 Fix workflow name Andrei Betlen 2023-04-09 22:38:19 -04:00
  • fda975e5a9 Rename test publish Andrei Betlen 2023-04-09 22:34:17 -04:00
  • baa394491c Add PyPI publish workflow Andrei Betlen 2023-04-09 22:32:30 -04:00
  • 8c2bb3042f Bump version Andrei Betlen 2023-04-09 22:12:23 -04:00
  • c3c2623e8b Update llama.cpp Andrei Betlen 2023-04-09 22:01:33 -04:00
  • e636214b4e Add test publish workflow Andrei Betlen 2023-04-08 19:57:37 -04:00
  • 314ce7d1cc Fix cpu count default Andrei Betlen 2023-04-08 19:54:04 -04:00
  • 3fbc06361f Formatting Andrei Betlen 2023-04-08 16:01:45 -04:00
  • 0067c1a588 Formatting Andrei Betlen 2023-04-08 16:01:18 -04:00
  • 0a5c551371 Bump version Andrei Betlen 2023-04-08 15:09:48 -04:00
  • 38f442deb0 Bugfix: Wrong size of embeddings. Closes #47 Andrei Betlen 2023-04-08 15:05:33 -04:00
  • 6d1bda443e Add clients example. Closes #46 Andrei Betlen 2023-04-08 09:35:32 -04:00
  • c940193e64 Bump version Andrei Betlen 2023-04-08 03:13:39 -04:00
  • edaaa1bd63 Only build wheels on workflow dispatch Andrei Betlen 2023-04-08 03:11:25 -04:00
  • ae3e9c3d6f Update shared library extension for macos Andrei Betlen 2023-04-08 02:45:21 -04:00
  • 6a143ac0db Merge branch 'main' of github.com:abetlen/llama_cpp_python into main Andrei Betlen 2023-04-08 02:40:42 -04:00
  • e611cfc56d Build shared library with make on unix platforms Andrei Betlen 2023-04-08 02:39:17 -04:00
  • a3f713039f Update llama.cpp Andrei Betlen 2023-04-08 02:38:42 -04:00
  • 41365b0456 Merge pull request #15 from SagsMug/main Andrei 2023-04-07 20:43:33 -04:00
  • 16fc5b5d23 More interoperability to the original llama.cpp, and arguments now work Mug 2023-04-07 13:32:19 +02:00
  • c3b1aa6ab7 Clone submodule Andrei Betlen 2023-04-07 03:19:07 -04:00
  • d4912a80da Install build dependencies Andrei Betlen 2023-04-07 03:18:56 -04:00
  • d74800da52 Build wheels Andrei Betlen 2023-04-07 03:14:38 -04:00
  • 0fd32046cb Bump version Andrei Betlen 2023-04-06 22:48:54 -04:00
  • 88c23d04a8 Fix windows dll location issue Andrei Betlen 2023-04-06 22:44:31 -04:00
  • 241722c981 Quote destination Andrei Betlen 2023-04-06 22:38:53 -04:00
  • d75196d7a1 Install with pip during build step Andrei Betlen 2023-04-06 21:33:54 -04:00
  • dd1c298620 Fix typo Andrei Betlen 2023-04-06 21:28:03 -04:00
  • baa825dacb Add windows and mac runners Andrei Betlen 2023-04-06 21:27:01 -04:00
  • da539cc2ee Safer calculation of default n_threads Andrei Betlen 2023-04-06 21:22:19 -04:00
  • 9b7526895d Bump version Andrei Betlen 2023-04-06 21:19:08 -04:00
  • 7851cc1e3c Don't install pydantic by default Andrei Betlen 2023-04-06 21:10:34 -04:00
  • 09707f5b2a Remove console script Andrei Betlen 2023-04-06 21:08:32 -04:00
  • 930db37dd2 Merge branch 'main' of github.com:abetlen/llama_cpp_python into main Andrei Betlen 2023-04-06 21:07:38 -04:00
  • 55279b679d Handle prompt list Andrei Betlen 2023-04-06 21:07:35 -04:00
  • c2e690b326 Merge pull request #29 from MillionthOdin16/main Andrei 2023-04-06 21:06:31 -04:00
  • 10c7571117 Fixed too many newlines, now onto args. Mug 2023-04-06 15:33:22 +02:00
  • 085cc92b1f Better llama.cpp interoperability Has some too many newline issues so WIP Mug 2023-04-06 15:30:57 +02:00
  • 2e91affea2 Ignore ./idea folder MillionthOdin16 2023-04-05 18:23:17 -04:00
  • c283edd7f2 Set n_batch to default values and reduce thread count: MillionthOdin16 2023-04-05 18:17:29 -04:00
  • b9b6dfd23f Merge remote-tracking branch 'origin/main' MillionthOdin16 2023-04-05 17:51:43 -04:00
  • 76a82babef Set n_batch to the default value of 8. I think this is leftover from when n_ctx was missing and n_batch was 2048. MillionthOdin16 2023-04-05 17:44:53 -04:00
  • 38f7dea6ca Update README and docs Andrei Betlen 2023-04-05 17:44:25 -04:00
  • 1e90597983 Add pydantic dep. Errors if pedantic isn't present. Also throws errors relating to TypeDict or subclass() if the version is too old or new... MillionthOdin16 2023-04-05 17:37:06 -04:00
  • 267d3648fc Bump version Andrei Betlen 2023-04-05 16:26:22 -04:00
  • 74bf043ddd Update llama.cpp Andrei Betlen 2023-04-05 16:25:54 -04:00
  • 44448fb3a8 Add server as a subpackage Andrei Betlen 2023-04-05 16:23:25 -04:00
  • e1b5b9bb04 Update fastapi server example Andrei Betlen 2023-04-05 14:44:26 -04:00
  • 283e59c5e9 Fix bug in init_break not being set when exited via antiprompt and others. Mug 2023-04-05 14:47:24 +02:00
  • 99ceecfccd Move to new examples directory Mug 2023-04-05 14:28:02 +02:00
  • e3ea354547 Allow local llama library usage Mug 2023-04-05 14:23:01 +02:00
  • e4c6f34d95 Merge branch 'main' of https://github.com/abetlen/llama-cpp-python Mug 2023-04-05 14:18:27 +02:00
  • 6de2f24aca Bump version Andrei Betlen 2023-04-05 06:53:43 -04:00
  • e96a5c5722 Make Llama instance pickleable. Closes #27 Andrei Betlen 2023-04-05 06:52:17 -04:00
  • 152e4695c3 Bump Version Andrei Betlen 2023-04-05 04:43:51 -04:00
  • c177c807e5 Add supported python versions Andrei Betlen 2023-04-05 04:43:19 -04:00
  • 17fdd1547c Update workflow name and add badge to README Andrei Betlen 2023-04-05 04:41:24 -04:00
  • 7643f6677d Bugfix for Python3.7 Andrei Betlen 2023-04-05 04:37:33 -04:00
  • 4d015c33bd Fix syntax error Andrei Betlen 2023-04-05 04:35:15 -04:00
  • 47570df17b Checkout submodules Andrei Betlen 2023-04-05 04:34:19 -04:00
  • e3f999e732 Add missing scikit-build install Andrei Betlen 2023-04-05 04:31:38 -04:00
  • 43c20d3282 Add initial github action to run automated tests Andrei Betlen 2023-04-05 04:30:32 -04:00
  • b1babcf56c Add quantize example Andrei Betlen 2023-04-05 04:17:26 -04:00
  • c8e13a78d0 Re-organize examples folder Andrei Betlen 2023-04-05 04:10:13 -04:00
  • c16bda5fb9 Add performance tuning notebook Andrei Betlen 2023-04-05 04:09:19 -04:00
  • cefc69ea43 Add runtime check to ensure embedding is enabled if trying to generate embeddings Andrei Betlen 2023-04-05 03:25:37 -04:00
  • 5c50af7462 Remove workaround Andrei Betlen 2023-04-05 03:25:09 -04:00
  • c3972b61ae Add basic tests. Closes #24 Andrei Betlen 2023-04-05 03:23:15 -04:00
  • 51dbcf2693 Bugfix: wrong signature for quantize function Andrei Betlen 2023-04-04 22:36:59 -04:00
  • 8279fb7d92 Bump version Andrei Betlen 2023-04-04 17:17:11 -04:00
  • c137789143 Add verbose flag. Closes #19 Andrei Betlen 2023-04-04 13:09:24 -04:00
  • 5075c16fcc Bugfix: n_batch should always be <= n_ctx Andrei Betlen 2023-04-04 13:08:21 -04:00
  • c862e8bac5 Fix repeating instructions and an antiprompt bug Mug 2023-04-04 17:54:47 +02:00
  • 248b0566fa Update README Andrei Betlen 2023-04-04 10:57:22 -04:00
  • 9cde7973cc Fix stripping instruction prompt Mug 2023-04-04 16:20:27 +02:00
  • da5a6a7089 Added instruction mode, fixed infinite generation, and various other fixes Mug 2023-04-04 16:18:26 +02:00
  • 0b32bb3d43 Add instruction mode Mug 2023-04-04 11:48:48 +02:00
  • ffe34cf64d Allow user to set llama config from env vars Andrei Betlen 2023-04-04 00:52:44 -04:00
  • 05eb2087d8 Small fixes for examples Andrei Betlen 2023-04-03 20:33:07 -04:00
  • caf3c0362b Add return type for default __call__ method Andrei Betlen 2023-04-03 20:26:08 -04:00
  • 4aa349d777 Add docstring for create_chat_completion Andrei Betlen 2023-04-03 20:24:20 -04:00
  • 4615f1e520 Add chat completion method to docs Andrei Betlen 2023-04-03 20:14:03 -04:00
  • 5cf29d0231 Bump version Andrei Betlen 2023-04-03 20:13:46 -04:00
  • 7fedf16531 Add support for chat completion Andrei Betlen 2023-04-03 20:12:44 -04:00
  • 3dec778c90 Update to more sensible return signature Andrei Betlen 2023-04-03 20:12:14 -04:00
  • f7ab8d55b2 Update context size defaults Close #11 Andrei Betlen 2023-04-03 20:11:13 -04:00
  • c0a5c0171f Add embed back into documentation Andrei Betlen 2023-04-03 18:53:00 -04:00
  • adf656d542 Bump version Andrei Betlen 2023-04-03 18:46:49 -04:00
  • ae004eb69e Fix #16 Andrei Betlen 2023-04-03 18:46:19 -04:00