vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Commit graph

  • a5c42c4b13
         Fix typo in llama.h (#593) master-a5c42c4 anzz1 2023-03-29 16:19:29 +03:00
  • 49c4c225b5 Merge branch 'master' into concedo Concedo 2023-03-29 21:08:03 +08:00
  • 271307232c Merged PR with a few changes: Concedo 2023-03-29 20:38:57 +08:00
  • b7a3365f4a
         Fix GCC warning about binary literal anzz1 2023-03-29 15:28:26 +03:00
  • 73071045c9
         Fix typo in llama.h anzz1 2023-03-29 14:22:50 +03:00
  • b3a360d80c
         Create chat-13B.bat Thérence 2023-03-29 10:06:56 +02:00
  • ff9b824a2a
         fixed whitespace in reverse prompt issue Tobias Lütke 2023-03-29 09:56:11 +02:00
  • 3f5f4286dd Always initialize mm_addr and mm_length in llama_model Slaren 2023-03-29 08:53:14 +02:00
  • baa529e9c0 Unmap the file in llama_free Slaren 2023-03-29 08:31:26 +02:00
  • e6f1c19937 Make mmap_file static Slaren 2023-03-29 06:18:18 +02:00
  • 7961493a40 Fix ggml_init_params in quantize Slaren 2023-03-29 05:38:57 +02:00
  • ef9afe1540 Add mmap support for model files Slaren 2023-03-29 02:03:43 +02:00
  • 13b4c05d66 Some more code cleanup InconsolableCellist 2023-03-28 16:59:27 -06:00
  • 1041ddb2cd apply PR suggestions Tristan Carel 2023-03-28 22:36:56 +02:00
  • 08121f3aa8 parallelize the quantization process Tristan Carel 2023-03-28 17:46:43 +02:00
  • 9a1ded757b
         plain __cpuid is enough here anzz1 2023-03-29 01:24:48 +03:00
  • bb54708e40
         CI: Re-enable AVX512 testing (Windows-MSVC) anzz1 2023-03-29 01:01:06 +03:00
  • c9c820ff36
         Added support for _POSIX_MAPPED_FILES if defined in source (#564) mmap CoderRC 2023-03-28 17:26:25 -04:00
  • 88c6535377
         spelling... Tobias Lütke 2023-03-28 23:20:11 +02:00
  • c6e8014062
         add example of re-act pattern Tobias Lütke 2023-03-28 23:06:33 +02:00
  • 5a5f8b1501
         Enable Fused-Multiply-Add (FMA) and F16C/CVT16 vector extensions on MSVC (#375) master-5a5f8b1 anzz1 2023-03-28 22:44:29 +03:00
  • f1217055ea
         CI: fix subdirectory path globbing (#546) master-f121705 anzz1 2023-03-28 22:43:25 +03:00
  • 13addf2a78 Merge branch 'concedo' of github.com:InconsolableCellist/llamacpp-for-kobold into concedo InconsolableCellist 2023-03-28 13:43:19 -06:00
  • f7c905b0d0 Minor overhaul of code: InconsolableCellist 2023-03-28 13:39:34 -06:00
  • 003365907d updating to version 17 of embedded koboldAI, and adding host address support InconsolableCellist 2023-03-28 13:39:10 -06:00
  • 8765be59a9
         Update build.yml anzz1 2023-03-28 22:07:08 +03:00
  • 38bc9cef4e
         Merge branch 'mmap' into mmap CoderRC 2023-03-28 15:03:48 -04:00
  • 7f4c5c6651
         llama : fix linkage with mingw (#551) master-7f4c5c6 anzz1 2023-03-28 21:23:09 +03:00
  • 2a98bc18ea
         ggml : add AVX2 implementation of quantize_row_q4_1 (#515) master-2a98bc1 slaren 2023-03-28 20:06:03 +02:00
  • 6ab328d88c
         Make quantize_row_q4_1 static slaren 2023-03-28 20:05:37 +02:00
  • d0aaff571c
         py : add temporary script to convert old ggml files to newer version (#539) master-d0aaff5 thement 2023-03-28 19:55:42 +02:00
  • d0330fd783
         py : add capabiliy to convert from ggml back to torch or hf format for further consumption/training/finetuning (#403) Tai Duc Nguyen 2023-03-28 13:51:29 -04:00
  • 41669f67d8 Actually use AVX2 Slaren 2023-03-28 19:45:59 +02:00
  • e29652996b Add AVX2 implementation of quantize_row_q4_1 Slaren 2023-03-28 19:41:28 +02:00
  • 1a5ee11377
         Restore old -std= flags Justine Tunney 2023-03-28 10:36:25 -07:00
  • 1631298475
         Remove -std=foo compiler flags Justine Tunney 2023-03-28 10:23:34 -07:00
  • 99c5b27654
         ggml : refactor quantized processing functions (#509) master-99c5b27 Stephan Walter 2023-03-28 17:13:01 +00:00
  • 1229722c61
         Merge branch 'master' into q-refactor Georgi Gerganov 2023-03-28 20:11:56 +03:00
  • a0c2401359
         ggml : minor Georgi Gerganov 2023-03-28 20:10:14 +03:00
  • cbddf4661b
         Get mmap() working with WIN32 MSVC Justine Tunney 2023-03-28 09:27:41 -07:00
  • 692ce3164e
         py : removed unused model variable and verified that the code functions correctly with vocab_only setting. Also confirmed that the code works as expected after running with reduced memory usage due to deletion of no-longer-needed variable. (#547) DooWoong Lee (David) 2023-03-29 02:02:34 +09:00
  • 96f9c0506f
         ci : make ctest verbose, hopefully we see what is wrong with the sanitizer master-96f9c05 Georgi Gerganov 2023-03-28 20:01:09 +03:00
  • d502bc7c9d
         tests : free llama context at the end of the test master-d502bc7 Georgi Gerganov 2023-03-28 19:51:55 +03:00
  • 436e561931
         all : be more strict about converting float to double (#458) master-436e561 Stephan Walter 2023-03-28 16:48:20 +00:00
  • 21e9ce7574
         perplexity : add <cmath> Georgi Gerganov 2023-03-28 19:40:01 +03:00
  • 20e1e84884
         deploy : add a Package.swift for SwiftPM support (#393) master-20e1e84 Jed Fox 2023-03-28 11:39:01 -05:00
  • 61733d3b49
         all : prefer float over double where appropriate Georgi Gerganov 2023-03-28 19:11:31 +03:00
  • e4881686b4
         Make WIN32 mmap() improvements (#341) oKatanaaa 2023-03-21 01:46:44 +04:00
  • f68345e9b1
         Fix softmax in perplexity.cpp Stephan Walter 2023-03-26 12:36:55 +02:00
  • 3a42193b3d
         Test equivalence of round, SILU implementations Stephan Walter 2023-03-25 17:00:29 +01:00
  • 54b75a77fb
         Be more strict about converting float to double Stephan Walter 2023-03-24 10:26:44 +01:00
  • c1f885067c
         ggml : introduce structs for the q4 data blocks (#356) master-c1f8850 Stephan Walter 2023-03-28 15:56:03 +00:00
  • 6a3b29a923
         ggml : rename quant struct variables + fix ARM_NEON Georgi Gerganov 2023-03-28 18:52:33 +03:00
  • e0670260fb
         gitignore : add "embedding" Georgi Gerganov 2023-03-28 18:34:35 +03:00
  • ce3f7adc85
         Fix linking on mingw32 anzz1 2023-03-28 18:14:04 +03:00
  • 28ba975aea
         Check the existence of f16_model_path_base in quantize.py (#574) dotpy314 2023-03-28 23:06:28 +08:00
  • 25248d7391 Use the same threshold for OpenBLAS and ggml thread limiting Maël Kerbiriou 2023-03-28 16:51:45 +02:00
  • 2e6c295bc7
         CMake: Add explicit F16C option (x86) anzz1 2023-03-28 17:43:32 +03:00
  • a6bdc47cba
         Fix usage of F16C intrinsics in AVX code (#563) master-a6bdc47 slaren 2023-03-28 16:26:55 +02:00
  • 40c8e68122 Check the existence of f16_model_path_base in quantize.py Jincheng Miao 2023-03-28 22:13:16 +08:00
  • 7b8dbcb78b
         main.cpp fixes, refactoring (#571) master-7b8dbcb anzz1 2023-03-28 17:09:55 +03:00
  • 51266e4ae7
         n_keep help update anzz1 2023-03-28 16:54:29 +03:00
  • ebf09a1919
         * -> & anzz1 2023-03-28 16:02:40 +03:00
  • fcabe9b8b2
         found this one on the floor anzz1 2023-03-28 15:49:40 +03:00
  • 021bdf237a
         main.cpp fixes, refactoring anzz1 2023-03-28 15:43:16 +03:00
  • 911782cfdd Use more accurate function names Slaren 2023-03-28 14:29:09 +02:00
  • 7c97743ea6 Fix linker error for tests kirillsurkov 2023-03-28 13:40:25 +03:00
  • bf30406f50 Merge branch 'master' into concedo Concedo 2023-03-28 17:13:38 +08:00
  • 99590bf992
         CI: github runner avx512f detection fix (windows) anzz1 2023-03-28 11:31:49 +03:00
  • 9e03cba624
         Merge branch 'master' into feat-instruct-cpp anzz1 2023-03-28 09:17:04 +03:00
  • 5ffcf7e2de
         Add instruct.cpp to Makefile anzz1 2023-03-28 09:14:09 +03:00
  • 4b8efff0e3
         Add embedding example to Makefile (#540) master-4b8efff RJ Adriaansen 2023-03-28 08:11:09 +02:00
  • 30317983c4
         Added support for _POSIX_MAPPED_FILES if defined in source CoderRC 2023-03-27 20:47:37 -04:00
  • ab6ac3d335 Fix usage of F16C intrinsics in AVX code Slaren 2023-03-27 23:33:49 +02:00
  • 926e49e78e
         Merge branch 'ggerganov:master' into master R.Kaufmann 2023-03-27 21:11:08 +02:00
  • 120bda4c18
         Update alpaca.sh anzz1 2023-03-27 17:22:15 +03:00
  • 4971df4f5c
         remove -ins --instruct cmdline arg anzz1 2023-03-27 17:06:03 +03:00
  • c8048f0cbd
         instruct.cpp, continue on empty line, endless instruct mode, refactors anzz1 2023-03-27 16:32:48 +03:00
  • 46ddbb22bf allow url params Concedo 2023-03-27 17:40:05 +08:00
  • 6cf5bde4c3
         Revert 7e53955 (#542) anzz1 2023-03-27 10:51:27 +03:00
  • b983f4fc70 Removed unused model variable and verified that the code functions correctly with vocab_only setting. Also confirmed that the code works as expected after running with reduced memory usage due to deletion of no-longer-needed variable. fritzprix 2023-03-27 14:35:02 +09:00
  • 7e5395575a
         Fix missing ggml link in cmake for examples/* on w64-mingw32 (#542) Marco Matthies 2023-03-27 06:55:26 +02:00
  • 1bf20a34b0
         CI: fix subdirectory path globbing anzz1 2023-03-27 07:52:56 +03:00
  • 13dedf2dbe Fix missing ggml link in cmake for examples/* on w64-mingw32 Marco Matthies 2023-03-27 01:29:19 +02:00
  • 58b3cbc20b
         Add embedding example to Makefile RJ Adriaansen 2023-03-26 23:34:42 +02:00
  • 53a187d2c0 Add script to convert old ggml files to newer version Jakub Horak 2023-03-26 22:47:45 +02:00
  • b8a80f9994 Add __SSE3__ macro for MSVC too because why not anzz1 2023-03-22 05:47:05 +02:00
  • 67ad7c445c MSVC cvt intrinsics anzz1 2023-03-22 04:36:29 +02:00
  • c43d45696f Enable F16C/CVT16 vector extensions on MSVC anzz1 2023-03-22 04:05:49 +02:00
  • 50ea0027a4 Enable Fused-Multiply-Add (FMA) instructions on MSVC anzz1 2023-03-22 03:46:48 +02:00
  • 3a6b10d22d Refactor quantized processing functions Stephan Walter 2023-03-25 18:40:28 +01:00
  • 34c1072e49
         ci: add debug build to sanitizer build matrix (#527) master-34c1072 Erik Scholz 2023-03-26 17:48:40 +02:00
  • d54bed0e37
         ci: add debug build to sanitzier build matrix Green Sky 2023-03-26 16:26:22 +02:00
  • 939ad2d3a5
         Fix undefined variables in debug build, remove unused variables (#531) master-939ad2d Stephan Walter 2023-03-26 15:34:02 +00:00
  • 8fca119295 Fix undefined variables in debug build, remove unused variables Stephan Walter 2023-03-26 17:16:15 +02:00
  • 35e76863db Treat empty line input as "no input" Jakub Horak 2023-03-26 16:50:02 +02:00
  • 8c2ec5e21d
         Add support for linux/arm64 platform during Docker Builds (#514) master-8c2ec5e Juan Calderon-Perez 2023-03-26 10:48:42 -04:00
  • 417bd2d677 Add backwards-compatibility for older model format Jakub Horak 2023-03-26 16:23:11 +02:00
  • 92d10215ec Introduce structs for the q4 data blocks Stephan Walter 2023-03-21 16:14:59 +01:00
  • b391579db9
         Update README and comments for standalone perplexity tool (#525) master-b391579 Stephan Walter 2023-03-26 13:14:01 +00:00