Commit graph

  • f05fa2a2b1 Merge remote-tracking branch 'upstream/master' into feature-convert-py-autogen-uuid brian khuu 2024-07-27 13:04:45 +10:00
  • 3fb690e91b convert*.py: inline source uuid generation approach brian khuu 2024-07-27 13:03:13 +10:00
  • 83e6a17ddf llama : fix session file loading Francis Couture-Harpin 2024-07-26 22:56:57 -04:00
  • 2c73791d62 refactoring: remove dup code hongruichen 2024-07-27 10:47:18 +08:00
  • 2b1f616b20
    ggml : reduce hash table reset cost (#8698) b3468 slaren 2024-07-27 04:41:55 +02:00
  • 18aa6654d5 refactoring: opt graph key gen hongruichen 2024-07-27 10:38:43 +08:00
  • ae5331d0cf GGML_ABORT use format string slaren 2024-07-27 02:23:44 +02:00
  • 3ed1bc09f5 GGML_ASSERT(false) -> GGML_ABORT("fatal error") slaren 2024-07-26 16:39:10 +02:00
  • dd84a88a58 fix unreachable code warnings after GGML_ASSERT(false) slaren 2024-07-26 02:27:58 +02:00
  • ac6ce60ec4 ggml : reduce hash table reset cost slaren 2024-03-18 15:31:21 +01:00
  • 3ab3eca372 abstract rpc server Wenjing Yu 2024-07-26 18:10:20 -07:00
  • da6c1f29f2
    Merge pull request #10 from zihaoccc/cleanup7 Zihao Chen 2024-07-26 17:47:43 -07:00
  • a53d266f7b big cleanup Wenjing Yu 2024-07-26 17:47:11 -07:00
  • 3ba087913f
    Merge pull request #9 from zihaoccc/cleanup6 Zihao Chen 2024-07-26 17:00:26 -07:00
  • 6ed7279adb remove benchmark Wenjing Yu 2024-07-26 16:59:48 -07:00
  • 0f2350dcb6
    Merge pull request #8 from zihaoccc/cleanup5 Zihao Chen 2024-07-26 16:51:16 -07:00
  • 5545e8ff92 remove batch-benched Wenjing Yu 2024-07-26 16:50:28 -07:00
  • 5630607f2b
    Merge pull request #7 from zihaoccc/cleanup4 Zihao Chen 2024-07-26 16:47:56 -07:00
  • 4b850f0ce4 remove batched Wenjing Yu 2024-07-26 16:47:33 -07:00
  • 715540a77b
    Merge pull request #6 from zihaoccc/cleanup3 Zihao Chen 2024-07-26 16:43:23 -07:00
  • bf621daa86 remove tests 2 Wenjing Yu 2024-07-26 16:42:54 -07:00
  • f7c0f9f576
    Merge pull request #5 from zihaoccc/cleanup2 Zihao Chen 2024-07-26 16:39:41 -07:00
  • 4810ab1aa1 remove tests Wenjing Yu 2024-07-26 16:38:13 -07:00
  • 0e5165b605
    Merge pull request #4 from zihaoccc/cleanup1 Zihao Chen 2024-07-26 16:35:22 -07:00
  • b76557d7c6 remove ci Wenjing Yu 2024-07-26 16:34:50 -07:00
  • c8b424fae5 llama : remove _context suffix for llama_data_context Francis Couture-Harpin 2024-07-26 19:06:37 -04:00
  • 90fd87df41 address comments jmorganca 2024-07-26 15:10:49 -07:00
  • 1a3a1b6d54 address comments jmorganca 2024-07-26 05:41:17 -04:00
  • 24540dd2bc Update convert_hf_to_gguf.py Jeffrey Morgan 2024-07-25 00:52:15 -04:00
  • e6bacb405a Add llama 3.1 rope scaling factors to llama conversion and inference jmorganca 2024-07-24 15:46:47 -04:00
  • 7ff30834e1
    Add option to keep output and embed tensors at f16 Robert Sinclair 2024-07-26 23:42:35 +03:00
  • ab8b748188 change to one time init to prevent performance drop domke 2024-07-27 02:13:24 +09:00
  • 0c491520a8 convert-*.py: Add source uuid generation brian khuu 2024-07-27 02:25:39 +10:00
  • be9a8c73a0 fix: suppress warning hongruichen 2024-07-26 23:07:25 +08:00
  • 47735cb589 fix: try fix error in 2nd run by appending dimension into graph key hongruichen 2024-07-26 23:03:09 +08:00
  • ee305cc171 refactoring: split qnn rpc buffer into dedicated class hongruichen 2024-07-26 22:33:30 +08:00
  • 65f7455cea Modify 2 notes caitianchi 2024-07-26 21:49:23 +08:00
  • f3d400dac0 remove uhd_image_embed caitianchi 2024-07-26 21:15:03 +08:00
  • c08a2e58c8
    common : add --no-warmup option for main/llama-cli Daniel Bevenius 2024-07-26 15:10:22 +02:00
  • 7cc64329d7 ggml : reading the runtime sve config of the cpu domke 2024-07-26 18:54:02 +09:00
  • 01245f5b16
    llama : fix order of parameters (#8706) b3467 Judd 2024-07-26 16:38:12 +08:00
  • f091a1614c fix order of parameters Judd 2024-07-26 15:57:32 +08:00
  • 201559d177 Merge branch 'master' of https://github.com/JoanFM/llama.cpp into feat-jina-embeddings-v2-zh Joan Martinez 2024-07-26 09:18:18 +02:00
  • 43eef8d287
    Merge branch 'ggerganov:master' into gguf-model-template Austin 2024-07-26 00:06:35 -04:00
  • e58d3cc279
    Merge branch 'ggerganov:master' into fix-vulkan-shader-warnings Austin 2024-07-26 00:06:23 -04:00
  • cddc899b85 llama : various integer type cast and format string fixes Francis Couture-Harpin 2024-07-25 22:58:20 -04:00
  • 9e22064a0d llama : fix uint64_t format type Francis Couture-Harpin 2024-07-25 22:49:14 -04:00
  • 8e39037b86 llama : refactor session file management Francis Couture-Harpin 2024-07-25 18:33:54 -04:00
  • cd78f93710
    Merge pull request #3 from zihaoccc/cleanup Zihao Chen 2024-07-25 16:10:46 -07:00
  • 7addbe3e9d remove baby-llama Wenjing Yu 2024-07-25 16:10:05 -07:00
  • 8fd767a557
    Merge branch 'ggerganov:master' into master Zihao Chen 2024-07-25 15:49:32 -07:00
  • 1cd7ac090b Reimplement 'collapsed' unicode categories: jaime-m-p 2024-07-26 00:43:43 +02:00
  • 8f7d56ec5b Add unicode_data helper functions jaime-m-p 2024-07-26 00:26:42 +02:00
  • 8c8e1afaae Fix codepoint_categ return types jaime-m-p 2024-07-26 00:18:16 +02:00
  • ecebfc0c71 Update unicode data: sorted whitespaces jaime-m-p 2024-07-26 00:16:24 +02:00
  • 01aec4a631
    server : add Speech Recognition & Synthesis to UI (#8679) b3466 Yaiko 2024-07-25 18:10:16 -04:00
  • 41cd47caab
    examples : export-lora : fix issue with quantized base models (#8687) b3465 Xuan Son Nguyen 2024-07-25 23:49:39 +02:00
  • 49ce0ab6d4
    ggml: handle ggml_init failure to fix NULL pointer deref (#8692) b3464 DavidKorczynski 2024-07-25 22:23:05 +01:00
  • fde72a9f86 push partial code here Yutong Dai 2024-07-25 21:15:07 +00:00
  • c5f12a189d server : add Speech Recognition & Synthesis to UI (fixes) Yaiko 2024-07-25 16:43:24 -04:00
  • 13b5bc3e83
    Install curl in runtime layer Brandon Squizzato 2024-07-25 16:42:12 -04:00
  • 0f9868e52c
    ggml: handle ggml_init failure to fix NULL pointer deref DavidKorczynski 2024-07-25 21:08:06 +01:00
  • 4226a8d10e
    llama : fix build + fix fabs compile warnings (#8683) b3463 Georgi Gerganov 2024-07-25 19:57:31 +03:00
  • bf5a81df37
    ggml : fix build on Windows with Snapdragon X (#8531) b3462 Andreas (Andi) Kunar 2024-07-25 18:01:00 +02:00
  • 88954f7fbd
    tests : fix printfs (#8068) b3461 Georgi Gerganov 2024-07-25 18:57:44 +03:00
  • f9e2dc00ac
    Merge branch 'ggerganov:master' into master MONONOKE 2024-07-25 23:45:04 +08:00
  • f843e5aaf5 fix: 1.free up rpc memory at destruct hongruichen 2024-07-22 23:41:23 +08:00
  • 53ad4bd89f Add support for loongarch backend in sgemm.cpp Tianzhengshuyuan 2024-07-25 22:32:41 +08:00
  • 28dd1b1f08
    reword error message sasha0552 2024-07-25 12:29:31 +00:00
  • 9aeb0e1f75 sycl add conv support sycl-conv-op Meng, Hengyu 2024-07-25 12:14:30 +00:00
  • e8f1bd8b39
    common : support for lifecycle scripts sasha0552 2024-07-24 15:46:04 +00:00
  • 65cf58ebe3 export-lora : fix issue with quantized base models ngxson 2024-07-25 14:01:10 +02:00
  • ed67bcb24f
    [SYCL] fix multi-gpu issue on sycl (#8554) b3460 Chen Xi 2024-07-25 11:45:18 +00:00
  • eddcb5238b
    ggml : add and use ggml_cpu_has_llamafile() (#8664) b3459 Georgi Gerganov 2024-07-25 12:37:42 +03:00
  • fc76684325 fix typo Chen Xi 2024-07-25 08:45:00 +00:00
  • 8fe8086363 add linux part change on doc Chen Xi 2024-07-25 08:42:03 +00:00
  • be6d7c0791
    examples : remove finetune and train-text-from-scratch (#8669) b3458 Xuan Son Nguyen 2024-07-25 10:39:04 +02:00
  • fcce8732a1 add doc for sycl multi-card Chen Xi 2024-07-25 08:29:06 +00:00
  • 4b0eff3df5
    docs : Quantum -> Quantized (#8666) vbatts-finetune Ujjawal Panchal 2024-07-25 13:43:27 +05:30
  • 72b962925b delete minicpmv-wrapper in pr caitianchi 2024-07-25 16:01:26 +08:00
  • 5f4de35753
    llama : fix build + fix fabs compile warnings Georgi Gerganov 2024-07-25 10:26:11 +03:00
  • 107e1edb20 fix uhd code for review comment caitianchi 2024-07-25 15:22:11 +08:00
  • 8a4bad50a8
    llama: use sliding window for phi3 (#8627) b3456 Fan Shupei 2024-07-25 15:21:09 +08:00
  • 10895d3589 server : add Speech Recognition & Synthesis to UI Yaiko 2024-07-24 21:09:27 -04:00
  • 5afc5cf842 Fix CI build failure Xiaodong Ye 2024-07-25 09:04:28 +08:00
  • 68504f0970
    readme : update games list (#8673) MorganRO8 2024-07-24 12:48:00 -04:00
  • 657faff9cc
    Update README.md MorganRO8 2024-07-24 12:09:28 -04:00
  • 38c7bf27f1 fix small typo for export-lora ngxson 2024-07-24 16:43:44 +02:00
  • 3c965b4f58 update help message ngxson 2024-07-24 16:25:26 +02:00
  • cedacf0092 fix build ngxson 2024-07-24 16:17:43 +02:00
  • 16c889bc00 examples : remove finetune and train-text-from-scratch ngxson 2024-07-24 15:56:29 +02:00
  • f19bf99c01
    Build Llama SYCL Intel with static libs (#8668) Joe Todd 2024-07-24 14:36:00 +01:00
  • 1208c44003 Build Llama SYCL Intel with static libs Joe Todd 2024-07-24 14:00:27 +01:00
  • 3a7ac5300a
    readme : update UI list [no ci] (#8505) Thorsten Sommer 2024-07-24 14:52:30 +02:00
  • 96952e7181
    llama : fix llama_chat_format_single for mistral (#8657) b3452 Xuan Son Nguyen 2024-07-24 13:48:46 +02:00
  • 79167d9e49
    Re-add erroneously removed -fsycl from GGML_EXTRA_LIBS (#8667) b3451 Joe Todd 2024-07-24 11:55:26 +01:00
  • 20f56a02db use printf ngxson 2024-07-24 12:17:33 +02:00
  • b115105f05
    add llama_lora_adapter_clear (#8653) b3450 Xuan Son Nguyen 2024-07-24 11:25:19 +02:00
  • 9ee5651433 Re-add erroneously removed -fsycl from GGML_EXTRA_LIBS Joe Todd 2024-07-24 10:14:27 +01:00
  • 5934580905
    ggml : add and use ggml_cpu_has_llamafile() gg/system-info-llamafile Georgi Gerganov 2024-07-24 11:31:41 +03:00