Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-20 01:31:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 368d0c8a9e Respect the maximum number of tokens in interactive. (#298) tjohnman 2023-03-19 19:31:17 +01:00
  • 50fae10d03 Add --ignore-eos parameter (#181) slaren 2023-03-19 19:22:48 +01:00
  • 084e2f0ec0 interactive mode: print '\n' in sigint_handler, this flush stdout thus ensure color reset. (#283) Qingyou Meng 2023-03-20 02:10:00 +08:00
  • 0b366e7357 Command line switch to use F16 for memory_k and memory_v (refactor of #154) (#294) Erik Scholz 2023-03-19 18:57:00 +01:00
  • 160bfb217d Update hot topics to mention Alpaca support Georgi Gerganov 2023-03-19 19:51:55 +02:00
  • c494ed5b94 Fix off-by-one bug (#115) Georgi Gerganov 2023-03-19 19:46:32 +02:00
  • c1c7026b47 Fix python stuff (#109) Georgi Gerganov 2023-03-19 19:33:18 +02:00
  • 474f760411 updated binaries Concedo 2023-03-20 01:19:15 +08:00
  • a097703ec4 Merge branch 'master' into concedo Concedo 2023-03-20 01:18:42 +08:00
  • 29054a2bee explicit buffer allocation from python Concedo 2023-03-20 01:18:34 +08:00
  • 467b149761 Refactoring convert-pth-to-ggml.py: more concise and readable (#109) qunash 2023-03-19 20:17:39 +03:00
  • 70f01cb863 Drop trailing new line from file prompts (#80) Georgi Gerganov 2023-03-19 19:04:44 +02:00
  • 356c1b87ba bugfixes and support for persistent states Concedo 2023-03-20 00:59:45 +08:00
  • a4e63b73df Add instruction for using Alpaca (#240) Georgi Gerganov 2023-03-19 18:49:50 +02:00
  • 9e1707218a Add "--instruct" argument for usage with Alpaca (#240) Georgi Gerganov 2023-03-19 18:37:02 +02:00
  • 22213a17b5 Change RMSNorm eps to 1e-6 (#173) Georgi Gerganov 2023-03-19 17:30:00 +02:00
  • f952b7c613 Removed junk, fixed some bugs and support dynamic number of sharded files Concedo 2023-03-19 11:13:00 +08:00
  • d7def1a752 Warn user if a context size greater than 2048 tokens is specified (#274) Ronsor 2023-03-18 17:10:47 -07:00
  • 6f61c18ec9 Fix typo in readme Pavol Rusnak 2023-03-18 22:39:46 +01:00
  • 1e5a6d088d Add note about Python 3.11 to readme Pavol Rusnak 2023-03-18 22:20:04 +01:00
  • 554b541521 Add memory/disk requirements to readme Pavol Rusnak 2023-03-18 21:58:46 +01:00
  • c21c89edca Update README.md v1.0.0 LostRuins 2023-03-19 00:50:03 +08:00
  • 42f307ef6a Update README.md LostRuins 2023-03-19 00:21:59 +08:00
  • 2b188521a1 Merge branch 'ggerganov:master' into concedo LostRuins 2023-03-19 00:20:09 +08:00
  • 5a6f3b01bd update readme Concedo 2023-03-19 00:19:34 +08:00
  • 0dc3ab930c Updated binaries Concedo 2023-03-19 00:09:00 +08:00
  • e3d85aa08b Merge branch 'master' into concedo Concedo 2023-03-19 00:07:32 +08:00
  • 2c8f870f53 Created a python bindings for llama.cpp and emulated a simple Kobold HTTP API Endpoint Concedo 2023-03-19 00:07:11 +08:00
  • d3f202d57b Remove unused code since n_vocab is model.hparams.n_vocab (#262) Alex Nguyen 2023-03-18 20:51:49 +07:00
  • e03e359730 fixed warning with std::ignore about unused function result (#151) Justin Suess 2023-03-18 07:44:09 -04:00
  • a81d0c2a17 Fix n^2 loop in tokenization (#254) Gary Linscott 2023-03-18 04:17:19 -07:00
  • b2de7f18df CI Improvements (#230) anzz1 2023-03-18 09:27:12 +02:00
  • a19b5a4adc Merge remote-tracking branch 'origin/master' into concedo Concedo 2023-03-18 10:52:54 +08:00
  • a292747893 Nix flake (#40) Niklas Korz 2023-03-17 23:03:48 +01:00
  • c9f670a177 Implement non-greedy tokenizer that tries to maximize token lengths (#242) thement 2023-03-17 21:05:58 +01:00
  • 4f54609110 Default to 4 threads (#243) Georgi Gerganov 2023-03-17 21:46:46 +02:00
  • e81b9c81c1 Update Contributing section Georgi Gerganov 2023-03-17 20:30:04 +02:00
  • 367946c668 Don't tell users to use a bad number of threads (#243) Stephan Walter 2023-03-17 17:47:35 +00:00
  • 6b0df5ccf3 add ptread link to fix cmake build under linux (#114) mmyjona 2023-03-18 00:38:24 +08:00
  • 2af23d3043 🚀 Dockerize llamacpp (#132) Bernat Vadell 2023-03-17 10:47:06 +01:00
  • 904d2a8d6a Q4_1 quantization (#193) Matvey Soloviev 2023-03-17 05:48:39 +01:00
  • 3d4854455c ban eos token Concedo 2023-03-17 11:01:02 +08:00
  • 27990d54ed minor change (+1 squashed commits) oKatanaaa 2023-03-16 22:22:42 +00:00
  • 197020deee Use F16 for memory_k and memory_v Ty Everett 2023-03-14 23:10:12 -07:00
  • 7b8858415e Scale buf_size linearly with n_ctx hx507 2023-03-17 05:11:49 +08:00
  • 721311070e Update README.md Georgi Gerganov 2023-03-16 15:00:09 +02:00
  • ac15de7895 Expand "Contributing" section Georgi Gerganov 2023-03-16 08:55:13 +02:00
  • 273abc47ff Update hot topics - RMSnorm Georgi Gerganov 2023-03-16 07:12:12 +02:00
  • 9b4a15b17d Fix RMS norm in GGML (#191) Nebula 2023-03-15 19:29:25 -04:00
  • 6eac39ba95 Add RMS norm and use it (#187) hoangmit 2023-03-15 18:41:38 -04:00
  • 27944c4206 fixed typo (#178) moritzbrantner 2023-03-15 21:35:25 +01:00
  • 2d15d6c9a9 add SIGINT support for _WIN32 environments (#120) Rickey Bowers Jr 2023-03-15 13:56:24 -06:00
  • 2d64715ad4 added ctx_size parameter (#148) Justin Suess 2023-03-15 15:42:40 -04:00
  • 16b2c61a22 fixed color reset on exit (#149) Justin Suess 2023-03-15 15:39:38 -04:00
  • 977295c700 Fix potential licensing issue (#126) Musab Gultekin 2023-03-15 22:39:06 +03:00
  • 956dfda8ad Use tokenizer.vocab_size() instead of hardcoding 32000 in convert-pth-to-ggml.py (#142) Ronsor 2023-03-15 12:37:50 -07:00
  • 113e685d18 inline -> static inline for "bytesFromNibbles" (#161) hoangmit 2023-03-15 15:05:14 -04:00
  • 47857e564c Don't use vdotq_s32 if it's not available (#139) Ronsor 2023-03-14 12:34:37 -07:00
  • 60f819a2b1 Add section to README on how to run the project on Android (#130) Radoslav Gerganov 2023-03-14 15:30:08 +02:00
  • 97ab2b2578 Add Misc section + update hot topics + minor fixes Georgi Gerganov 2023-03-14 09:43:52 +02:00
  • 2f700a2738 Add windows to the CI (#98) Sebastián A 2023-03-13 17:29:10 -03:00
  • c09a9cfb06 CMake build in Release by default (#75) Georgi Gerganov 2023-03-13 21:22:15 +02:00
  • 7ec903d3c1 Update contribution section, hot topics, limitations, etc. Georgi Gerganov 2023-03-13 19:21:51 +02:00
  • 4497ad819c Print system information Georgi Gerganov 2023-03-13 19:15:08 +02:00
  • ed6849cc07 Initial support for CMake (#75) Sebastián A 2023-03-13 14:12:33 -03:00
  • 41be0a3b3d Add NetBSD support. (#90) Thomas Klausner 2023-03-13 17:40:54 +01:00
  • 671d5cac15 Use fprintf for diagnostic output (#48) Pavol Rusnak 2023-03-13 17:39:56 +01:00
  • 84d9015c4a Use vdotq_s32 to improve performance (#67) Georgi Gerganov 2023-03-13 18:36:44 +02:00
  • 63fd76fbb0 Reduce model loading time (#43) uint256_t 2023-03-14 01:33:43 +09:00
  • 2a20f48efa Fix UTF-8 handling (including colors) (#79) Val Kharitonov 2023-03-13 12:24:18 -04:00
  • d1f224712d Add quantize script for batch quantization (#92) Pavol Rusnak 2023-03-13 17:15:20 +01:00
  • 1808ee0500 Add initial contribution guidelines Georgi Gerganov 2023-03-13 09:42:26 +02:00
  • a169bb889c Gate signal support on being on a unixoid system. (#74) Matvey Soloviev 2023-03-13 04:08:01 +01:00
  • 460c482540 Fix token count accounting Matvey Soloviev 2023-03-13 00:35:51 +01:00
  • c80e2a8f2a Revert "10% performance boost on ARM" Georgi Gerganov 2023-03-13 01:28:08 +02:00
  • 54a0e66ea0 Check for vdotq_s32 availability Georgi Gerganov 2023-03-13 01:21:03 +02:00
  • 543c57e991 Ammend to previous commit - forgot to update non-QRDMX branch Georgi Gerganov 2023-03-13 01:05:24 +02:00
  • 113a9e83eb 10% performance boost on ARM Georgi Gerganov 2023-03-13 00:56:10 +02:00
  • 404fac0d62 Fix color getting reset before prompt output done (#65) Matvey Soloviev 2023-03-12 23:07:34 +01:00
  • 1a0a74300f Update README.md Georgi Gerganov 2023-03-12 23:39:01 +02:00
  • 96ea727f47 Add interactive mode (#61) Matvey Soloviev 2023-03-12 22:13:28 +01:00
  • 9661954835 Fix typo in README (#45) Marc Köhlbrugge 2023-03-13 03:30:08 +07:00
  • f385f8dee8 Allow using prompt files (#59) Ben Garney 2023-03-12 13:28:36 -07:00
  • 02f0c6fe7f Add back top_k (#56) beiller 2023-03-12 16:23:15 -04:00
  • eb062bb012 Windows fixes (#31) Sebastián A 2023-03-12 17:15:00 -03:00
  • 7027a97837 Update README.md Georgi Gerganov 2023-03-12 22:09:26 +02:00
  • 2d555e5b42 Add CI (#60) Georgi Gerganov 2023-03-12 22:08:24 +02:00
  • 7c9e54e55e Revert "weights_only" arg - this causing more trouble than help Georgi Gerganov 2023-03-12 20:59:01 +02:00
  • b9bd1d0141 python/pytorch compat notes (#44) Oleksandr Nikitin 2023-03-12 14:16:33 +02:00
  • 129c7d1ea8 Add repetition penalty (#20) beiller 2023-03-12 05:27:42 -04:00
  • 702fddf5c5 Clarify meaning of hacking Georgi Gerganov 2023-03-12 09:03:25 +02:00
  • 7d86e25bf6 README: add "Supported platforms" + update hot topics Georgi Gerganov 2023-03-12 08:41:54 +02:00
  • a93120236f use weights_only in conversion script (#32) deepdiffuser 2023-03-11 22:36:35 -08:00
  • 6a9a67f0be Add LICENSE (#21) Pavol Rusnak 2023-03-12 07:36:03 +01:00
  • da1a4ff01f Update README.md Georgi Gerganov 2023-03-12 01:26:32 +02:00
  • 6b2cb6302f Fix a typo in model name (#16) Juraj Bednar 2023-03-11 18:32:20 +01:00
  • 4235e3d5b3 Update README.md Georgi Gerganov 2023-03-11 18:10:18 +02:00
  • f1eaff4721 Add AVX2 support for x86 architectures thanks to @Const-me ! Georgi Gerganov 2023-03-11 17:58:18 +02:00
  • a9e58529ea Fix un-initialized FP16 tables on x86 (#15, #2) Georgi Gerganov 2023-03-11 17:40:14 +02:00
  • 7d9ed7b25f Bump memory buffer Georgi Gerganov 2023-03-11 12:44:21 +02:00