Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 05:48:47 +01:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • a19b5a4adc Merge remote-tracking branch 'origin/master' into concedo Concedo 2023-03-18 10:52:54 +0800
  • b10383fbff merge oKatanaaa 2023-03-18 02:02:02 +0000
  • a292747893
         Nix flake (#40) Niklas Korz 2023-03-17 23:03:48 +0100
  • c9f670a177
         Implement non-greedy tokenizer that tries to maximize token lengths (#242) thement 2023-03-17 21:05:58 +0100
  • 4f54609110
         Default to 4 threads (#243) Georgi Gerganov 2023-03-17 21:46:46 +0200
  • e81b9c81c1
         Update Contributing section Georgi Gerganov 2023-03-17 20:30:04 +0200
  • d0f855a8df
         Update README.md Kevin Kwok 2023-03-17 11:13:47 -0700
  • 367946c668
         Don't tell users to use a bad number of threads (#243) Stephan Walter 2023-03-17 17:47:35 +0000
  • 7566d1ab9b Insert single space in front of the prompt Jakub Horak 2023-03-17 18:08:03 +0100
  • f60e972eb5
         better CI naming convention anzz1 2023-03-17 19:44:02 +0200
  • 7e1041a730 Implement non-greedy tokenizer that tries to maximize token lengths Jakub Horak 2023-03-17 17:35:41 +0100
  • 6adc143e31 Don't tell users to use a bad number of threads Stephan Walter 2023-03-17 18:27:22 +0100
  • e7bdee6cc9
         CI fine tuning anzz1 2023-03-17 19:17:59 +0200
  • 1e82fa82c6
         Merge pull request #26 from mcmonkey4eva/master Kevin Kwok 2023-03-17 09:44:40 -0700
  • 71d097898d
         Merge pull request #12 from anzz1/ci_test Kevin Kwok 2023-03-17 09:44:26 -0700
  • 197df5f096
         Merge pull request #18 from jxy/limit_tokens Kevin Kwok 2023-03-17 09:44:12 -0700
  • 6b0df5ccf3
         add ptread link to fix cmake build under linux (#114) mmyjona 2023-03-18 00:38:24 +0800
  • c6a0f83b12
         Docker image shouldn't be published on PRs Sebastián A 2023-03-17 13:36:28 -0300
  • c0f6681694
         Merge branch 'master' into fix-cmake-pthread Sebastián A 2023-03-17 13:33:42 -0300
  • af02c94798 add easy Windows install instructions to the readme Alex "mcmonkey" Goodwin 2023-03-17 04:49:41 -0700
  • 584add5c18 fix publish permission Bernat Vadell 2023-03-17 10:54:21 +0100
  • 2af23d3043
         🚀 Dockerize llamacpp (#132) Bernat Vadell 2023-03-17 10:47:06 +0100
  • 79a48d9876 feat: include all-in-one command tool & update readme.md Bernat Vadell 2023-03-17 10:31:53 +0100
  • fb324e035a improvement(tools): add example tpoisonooo 2023-03-17 16:56:45 +0800
  • 3c7cb413fb improvement(tools): optimize with argparse tpoisonooo 2023-03-17 16:53:53 +0800
  • 7f2a53363c
         CI Improvements anzz1 2023-03-17 10:08:49 +0200
  • 50fa1a006e Merge branch 'master' into feat/dockerize Bernat Vadell 2023-03-17 09:02:11 +0100
  • db4945c22a fix(ggml): compile error on centos 7.9 and gcc7 tpoisonooo 2023-03-17 15:06:09 +0800
  • 3877722a08 fix compile error on centos 7.5 chenliangyu 2023-03-17 15:00:12 +0800
  • 5be098f51e Compute remaining tokens along the way and exit if over Xiao-Yong Jin 2023-03-17 00:20:24 -0500
  • 5b8023d935
         Implement prototype for instant mmap() loading Justine Tunney 2023-03-16 22:03:09 -0700
  • 904d2a8d6a
         Q4_1 quantization (#193) Matvey Soloviev 2023-03-17 05:48:39 +0100
  • f8db3d6cd9 Improved quantize script SuajCarrot 2023-03-16 21:15:41 -0600
  • 3d4854455c ban eos token Concedo 2023-03-17 11:01:02 +0800
  • 27990d54ed minor change (+1 squashed commits) oKatanaaa 2023-03-16 22:22:42 +0000
  • 197020deee Use F16 for memory_k and memory_v Ty Everett 2023-03-14 23:10:12 -0700
  • 07d2da0cb0 forgot the newline that needs to be sent to the model Rickey Bowers Jr 2023-03-16 20:54:06 -0600
  • 8470458b64 fix coloring of last n_batch of prompt, and refactor line input Rickey Bowers Jr 2023-03-16 20:23:32 -0600
  • e8fc1d5491
         Merge branch 'fix-cmake-pthread' of https://github.com/mmyjona/llama.cpp into fix-cmake-pthread mmyjona 2023-03-17 10:15:35 +0800
  • 88f776b972
         separate make and cmake workflow mmyjona 2023-03-17 10:15:06 +0800
  • 376a4a2ab0
         Merge branch 'master' into ci_test anzz1 2023-03-17 04:11:13 +0200
  • 8a3d6f2ea4
         CI: Enable Linux/MacOS builds anzz1 2023-03-17 04:00:40 +0200
  • f2eda96137
         Fix Makefile anzz1 2023-03-17 03:58:34 +0200
  • 235a4115df
         Update build.yml Kevin Kwok 2023-03-16 18:39:41 -0700
  • 72f9fbe4d1
         Fix #2 Kevin Kwok 2023-03-16 18:36:09 -0700
  • 2af2331add
         Merge pull request #3 from bigattichouse/master Kevin Kwok 2023-03-16 18:30:17 -0700
  • bf244623ef
         Merge pull request #5 from anzz1/master Kevin Kwok 2023-03-16 18:26:03 -0700
  • b4db6b7235 makefile: Fix CPU feature detection on Haiku Alexander von Gluck IV 2023-03-16 20:20:01 -0500
  • fc924e9aba minor change oKatanaaa 2023-03-16 22:38:06 +0000
  • 7252a2b658 refactor: make weights load faster oKatanaaa 2023-03-16 22:22:42 +0000
  • c2577fdb04
         Merge branch 'master' into master qunash 2023-03-17 00:31:52 +0300
  • 0660aeed69 process the scanf() output so Ubuntu 22 compiler doesn't error due to default warn_unused_result instead of Makefile -Wunused-result Michael E. Johnson 2023-03-16 16:18:12 -0500
  • 7b8858415e
         Scale buf_size linearly with n_ctx hx507 2023-03-17 05:11:49 +0800
  • c8917ca358
         Fix Windows CI and autorelease anzz1 2023-03-16 23:08:18 +0200
  • 6d973a699f
         Fix CMake anzz1 2023-03-16 23:07:23 +0200
  • 9f90347596
         Update main.cpp Justin Suess 2023-03-16 12:45:24 -0400
  • e750735216 first commit antimatter15 2023-03-16 05:56:54 -0700
  • 721311070e
         Update README.md Georgi Gerganov 2023-03-16 15:00:09 +0200
  • 0bc1e806d1 fix docker.yml Bernat Vadell 2023-03-16 12:02:22 +0100
  • 4941df7f3b fix github action docker Bernat Vadell 2023-03-16 12:00:05 +0100
  • c6b2c6fd8f include docker versioned images Bernat Vadell 2023-03-16 11:44:00 +0100
  • c20281924c fix name runs-on macOS-latest is macos-latest (lowercase) Bernat Vadell 2023-03-16 11:43:49 +0100
  • 3bcfc2b83d change CI to publish at github docker registry Bernat Vadell 2023-03-16 11:31:49 +0100
  • 60cf70725e Merge branch 'master' into feat/dockerize Bernat Vadell 2023-03-16 11:23:07 +0100
  • 2e2f144bcd
         Update README.md Geraint Ballinger 2023-03-16 09:07:10 +0000
  • ac15de7895
         Expand "Contributing" section Georgi Gerganov 2023-03-16 08:55:13 +0200
  • fcafaf03b0
         Add files via upload Black-Engineer 2023-03-16 08:23:02 +0200
  • 273abc47ff
         Update hot topics - RMSnorm Georgi Gerganov 2023-03-16 07:12:12 +0200
  • 610719ecc8
         Merge branch 'ggerganov:master' into fix-cmake-pthread mmyjona 2023-03-16 10:25:05 +0800
  • 7b051e3fb0
         add cmake to linux and macos platform mmyjona 2023-03-16 10:23:09 +0800
  • e13991e7cc Fix non-vectorised q4_1 vec mul Matvey Soloviev 2023-03-16 00:36:49 +0100
  • edbd4a0534 Fix ggml_vec_mad_q4_1 too Matvey Soloviev 2023-03-16 00:23:42 +0100
  • a2e9d4951b Rearrange Q4_1 quantization to work for multipart models. (Fix #152) Matvey Soloviev 2023-03-15 23:54:46 +0100
  • ad051ac56e Small optimisations to q4_1 dot product (@Const-me) Matvey Soloviev 2023-03-15 01:29:36 +0100
  • 561733cafa Add AVX2 version of ggml_vec_dot_q4_1 Matvey Soloviev 2023-03-14 19:37:58 +0100
  • 9b4a15b17d
         Fix RMS norm in GGML (#191) Nebula 2023-03-15 19:29:25 -0400
  • fa2aa1728a
         Fix RMS norm in GGML nebulatgs 2023-03-15 19:04:08 -0400
  • 68e421da89
         Grammar improvements and more information glencoe2004 2023-03-15 18:54:50 -0400
  • 599b3ca1e7
         Add files via upload glencoe2004 2023-03-15 18:53:24 -0400
  • 6eac39ba95
         Add RMS norm and use it (#187) hoangmit 2023-03-15 18:41:38 -0400
  • 11e594884b
         update op num hoangmit 2023-03-15 18:29:32 -0400
  • a8f75ec41e
         add ggml_rms_norm hoangmit 2023-03-15 18:26:19 -0400
  • d70f1fc72f
         Add files via upload glencoe2004 2023-03-15 18:23:48 -0400
  • 11b3dafd2c
         Made the file clearer for future python versions Reelix 2023-03-15 23:36:55 +0200
  • ae43513b39
         Added install instructions for latest python3 Reelix 2023-03-15 23:32:01 +0200
  • 5f0ada21bf fix Dmitry Wolf 2023-03-16 00:30:07 +0300
  • b430c51f99 Add --ignore-eos parameter Slaren 2023-03-15 22:14:09 +0100
  • 27944c4206
         fixed typo (#178) moritzbrantner 2023-03-15 21:35:25 +0100
  • bb034e5f78
         fixed typo moritzbrantner 2023-03-15 21:29:56 +0100
  • de17acac72 readme Dmitry Wolf 2023-03-15 23:24:48 +0300
  • c758d3d521 Merge branch 'master' into streaming Dmitry Wolf 2023-03-15 23:19:52 +0300
  • 0deb075a3f clean Dmitry Wolf 2023-03-15 22:57:59 +0300
  • 2d15d6c9a9
         add SIGINT support for _WIN32 environments (#120) Rickey Bowers Jr 2023-03-15 13:56:24 -0600
  • 2d64715ad4
         added ctx_size parameter (#148) Justin Suess 2023-03-15 15:42:40 -0400
  • 16b2c61a22
         fixed color reset on exit (#149) Justin Suess 2023-03-15 15:39:38 -0400
  • 977295c700
         Fix potential licensing issue (#126) Musab Gultekin 2023-03-15 22:39:06 +0300
  • bb997650be separate threads for r/w ops Dmitry Wolf 2023-03-15 22:38:28 +0300
  • 956dfda8ad
         Use tokenizer.vocab_size() instead of hardcoding 32000 in convert-pth-to-ggml.py (#142) Ronsor 2023-03-15 12:37:50 -0700
  • b4e2723a67
         Update main.cpp Georgi Gerganov 2023-03-15 21:36:58 +0200
  • f056beb384
         Apply suggestions from code review Georgi Gerganov 2023-03-15 21:35:48 +0200