Commit Graph

  • ffda777993
    ggml : add check for grad_accs (ggml/1046) Daniel Bevenius 2024-12-13 08:19:38 +0100
  • 8541e2cf9e
    common : remove old types Georgi Gerganov 2024-12-10 17:19:09 +0200
  • ce2b75d2fb
    CUDA: fix shared memory access condition for mmv (llama/10740) Johannes Gäßler 2024-12-09 20:07:12 +0100
  • 37df308a2a
    vulkan: fix compile warnings (llama/10731) Jeff Bolz 2024-12-09 01:24:01 -0600
  • 0e4eb51283
    Vulkan: fix NaN in tanh.comp with AMD proprietary driver on Windows (llama/10723) stduhpf 2024-12-08 19:19:19 +0100
  • 42a5245e6d
    vulkan: compile a test shader in cmake to check for coopmat2 support (llama/10713) Jeff Bolz 2024-12-08 02:05:55 -0600
  • 349430cce5
    ggml : disable iq4_nl interleave size 8 (llama/10709) Georgi Gerganov 2024-12-07 18:38:15 +0200
  • f4a6693239
    ggml : refactor online repacking (llama/10446) Djip007 2024-12-07 13:37:50 +0100
  • 5e59b3f6da
    Vulkan: VK_KHR_cooperative_matrix support to speed up prompt processing (llama/10597) 0cc4m 2024-12-07 10:24:15 +0100
  • d72153e3f7
    metal : Extend how Llama.cpp locates metal resources (llama/10676) Robert Ormandi 2024-12-07 01:55:01 -0600
  • b0dd14aaea
    vulkan: Add VK_NV_cooperative_matrix2 support for mul_mat and flash attention (llama/10206) Jeff Bolz 2024-12-05 13:15:05 -0600
  • f897eb7670
    whisper : support no_speech_thold (#2625) Karthick 2024-12-17 22:45:47 +0530
  • 2f2841bfce
    whisper : add single-timestamp logic (#2629) Karthick 2024-12-17 22:37:08 +0530
  • 09a1b61218
    readme : fix typo (#2637) crummyh 2024-12-17 11:05:35 -0600
  • 94e7da1ff2
    cmake : fix "amd64" processor string (#2638) Georgi Gerganov 2024-12-17 18:34:32 +0200
  • 47dfb9e265
    cmake : fix "amd64" processor string #2638 Georgi Gerganov 2024-12-17 18:33:28 +0200
  • 4dfc72e89e
    Fix typo in Java Binding README #2637 crummyh 2024-12-17 10:17:17 -0600
  • c4aed6831e
    vulkan : fix soft_max.comp division by zero (#2633) gn64 2024-12-16 19:34:38 +0900
  • ff67d10f1c
    ci : msys enable SDL2 build #2635 Georgi Gerganov 2024-12-16 08:59:45 +0200
  • 199579652e
    common : add cstdio header Georgi Gerganov 2024-12-16 08:57:04 +0200
  • 2fe659b844
    Accept review comments related to formatting. #2629 Karthick 2024-12-16 10:58:48 +0530
  • ec05d7705a allow stream prompt #2634 wznmickey 2024-12-15 23:03:35 -0500
  • d17e7139d8
    stream : update build instructions Georgi Gerganov 2024-12-15 21:55:36 +0200
  • 0ac64a321b
    Merge 36dcb97180 into 6a52eaea74 #2068 mytang0 2024-12-15 14:27:13 +0100
  • 6de6ca83fd add utf-8 encoding #2604 Hidetoshi Matsuo 2024-12-15 09:23:20 +0900
  • a120fc14d9 Fix hallucinations during silence Karthick Jeyapal 2024-12-14 23:23:24 +0530
  • 55fc15d1d8
    Merge d029175359 into 6a52eaea74 #1086 byte-6174 2024-12-14 16:56:17 +0100
  • 6a52eaea74
    android : fix build and ci (#2624) Thamster 2024-12-14 10:25:53 -0500
  • 31edfdbd32
    Merge 15b2f9bb08 into 6aa1d7b892 #1164 Alex Young 2024-12-13 14:54:05 -0800
  • 1be4938ae2 attempt to re-enable CI for JNI android #2624 Your Name 2024-12-13 13:11:25 -0500
  • 3448759086 Addressed review comments #2625 Karthick Jeyapal 2024-12-13 20:13:53 +0530
  • 72c277ffa8 Implement no_speech_thold Karthick Jeyapal 2024-12-13 10:56:23 +0530
  • 8e5fd65cd9 add openvino outputs Hidetoshi Matsuo 2024-12-13 14:29:00 +0900
  • faffedb3e7 Adding missing CMakeLists.txt include for ggm-cpu needed by whisper.android Your Name 2024-12-12 12:25:43 -0500
  • 6aa1d7b892
    models : fix typo in download-ggml-model.sh (#2623) Michael Rienstra 2024-12-12 08:02:00 -0800
  • adb75eea35
    Merge 12c9111fba into 262e865a70 #1455 Tyler 2024-12-12 14:26:20 +0100
  • 49c33aa40d
    Fix typo in download-ggml-model.sh #2623 Michael Rienstra 2024-12-11 13:49:59 -0800
  • 262e865a70
    ruby : Sync whisper.cpp and model download feature (#2617) KITAITI Makoto 2024-12-09 20:17:50 +0900
  • d4e47945e3 Use conditional get when get model files #2617 Kitaiti Makoto 2024-12-03 23:15:15 +0900
  • a0f3d8a831 Cosmetic fix Kitaiti Makoto 2024-12-01 08:26:47 +0900
  • 4559a70035 Don't care about no longer included file Kitaiti Makoto 2024-12-01 08:24:49 +0900
  • b8a5c85780 Remove unused function Kitaiti Makoto 2024-12-01 08:04:27 +0900
  • 0ed5b2399c Add headings to API section in README [skip ci] Kitaiti Makoto 2024-12-01 02:24:03 +0900
  • 9e50697dc1 Update documents Kitaiti Makoto 2024-12-01 02:10:01 +0900
  • d8d89d73e4 Add shorthand for pre-converted models Kitaiti Makoto 2024-12-01 02:09:44 +0900
  • 3fd13ae71f Make Whisper::Context#initialize accept Pathname Kitaiti Makoto 2024-11-29 22:09:55 +0900
  • d862e8359c Add test for Pathname of model Kitaiti Makoto 2024-11-29 22:05:52 +0900
  • b53b44e0ff Use C++17 Kitaiti Makoto 2024-12-06 23:05:42 +0900
  • ed733e85a1
    scripts : update to new build system v1.7.3-pre Georgi Gerganov 2024-12-09 11:30:16 +0200
  • 5980b1ae77
    devops : add cmake Georgi Gerganov 2024-12-08 23:09:26 +0200
  • 0415a66044
    devops : update make commands Georgi Gerganov 2024-12-08 23:07:29 +0200
  • 7d134e3737
    ggml : remove old files (skip) (#0) Georgi Gerganov 2024-12-08 23:04:26 +0200
  • 9df53b357e
    ggml : sync remnants (skip) (#0) Georgi Gerganov 2024-12-08 22:48:25 +0200
  • b2115b4d9b
    scripts : remove amx from sync Georgi Gerganov 2024-12-08 22:48:14 +0200
  • 0164427dd5 ci : disable freeBSD builds [no ci] Georgi Gerganov 2024-12-08 15:52:57 +0200
  • 627b11c78a readme : update build instructions Georgi Gerganov 2024-12-08 15:48:14 +0200
  • 472464453d ci : disable CUDA and Android builds Georgi Gerganov 2024-12-08 15:36:01 +0200
  • 11dddfbc9e ci : disable Obj-C build + fixes Georgi Gerganov 2024-12-08 13:35:35 +0200
  • 384e214cc7 make : shim cmake Georgi Gerganov 2024-12-06 15:34:53 +0200
  • f2c680f893 talk-llama : sync llama.cpp Georgi Gerganov 2024-12-05 14:30:33 +0200
  • fbe66da0e5 sync : ggml Georgi Gerganov 2024-12-05 14:29:18 +0200
  • a815940e0e ggml : add predefined list of CPU backend variants to build (llama/10626) Diego Devesa 2024-12-04 14:45:40 +0100
  • 904e307bce ggml-cpu : fix HWCAP2_I8MM value (llama/10646) Diego Devesa 2024-12-04 14:40:44 +0100
  • 491ec076b4 vulkan: Implement "fast divide" (mul+shift) for unary ops like copy (llama/10642) Jeff Bolz 2024-12-04 01:28:59 -0600
  • 966433fdf2 SYCL : Move to compile time oneMKL interface backend selection for NVIDIA backend (llama/10584) Nicolò Scipione 2024-12-04 02:29:20 +0100
  • 6f1ba9d82d Avoid using __fp16 on ARM with old nvcc (llama/10616) Frankie Robertson 2024-12-04 02:41:37 +0200
  • 015ecd0001 vulkan: optimize and reenable split_k (llama/10637) Jeff Bolz 2024-12-03 13:29:54 -0600
  • b7c64a4352 ggml: add GGML_SET Metal kernel + i32 CPU kernel (ggml/1037) PAB 2024-12-04 09:19:30 +0100
  • 7895d39508 ggml : add GGML_PAD_REFLECT_1D operation (ggml/1034) PAB 2024-12-03 20:20:04 +0100
  • 22616f00f9 files : remove make artifacts Georgi Gerganov 2024-12-03 20:29:32 +0200
  • 02c6fcbc2c common : fix compile warning Georgi Gerganov 2024-12-03 20:25:37 +0200
  • 3daeacad24 ggml : move AMX to the CPU backend (llama/10570) Diego Devesa 2024-12-03 20:22:12 +0200
  • 4d73962da4 metal : small-batch mat-mul kernels (llama/10581) Georgi Gerganov 2024-12-03 11:52:33 +0200
  • 068812650e SYCL: Fix and switch to GGML_LOG system instead of fprintf (llama/10579) Akarshan Biswas 2024-12-02 12:34:11 +0530
  • 4b7e059e15 ggml-cpu: replace AArch64 NEON assembly with intrinsics in ggml_gemv_q4_0_4x4_q8_0() (llama/10567) Adrien Gallouët 2024-11-30 18:13:18 +0100
  • 30e35d7271 vulkan: Dynamic subgroup size support for Q6_K mat_vec (llama/10536) Eve 2024-11-30 07:00:02 +0000
  • 3623bd58f2 ggml : fix I8MM Q4_1 scaling factor conversion (llama/10562) Georgi Gerganov 2024-11-29 16:25:39 +0200
  • cb847c20a7 ggml-cpu: fix typo in gemv/gemm iq4_nl_4_4 (llama/10580) Shupei Fan 2024-11-29 21:49:02 +0800
  • 964b154a2a sycl : offload of get_rows set to 0 (llama/10432) Alberto Cabrera Pérez 2024-11-29 12:38:45 +0000
  • d7c2a04bce sycl : Reroute permuted mul_mats through oneMKL (llama/10408) Alberto Cabrera Pérez 2024-11-29 09:49:43 +0000
  • 2bb4ca9cba CANN: RoPE operator optimization (llama/10563) Chenguang Li 2024-11-29 14:46:55 +0800
  • a753a82462 vulkan: get the first command buffer submitted sooner (llama/10499) Jeff Bolz 2024-11-29 00:18:02 -0600
  • 276b08d8f0 ggml : remove redundant copyright notice + update authors Georgi Gerganov 2024-11-28 20:46:40 +0200
  • 4ca1e72fe0 ggml : fix row condition for i8mm kernels (llama/10561) Georgi Gerganov 2024-11-28 14:56:37 +0200
  • 16a66f103f cmake : fix ARM feature detection (llama/10543) Georgi Gerganov 2024-11-28 14:56:23 +0200
  • 330273901f ggml-cpu: support IQ4_NL_4_4 by runtime repack (llama/10541) Shupei Fan 2024-11-28 20:52:03 +0800
  • 42099a9342 kompute : improve backend to pass test_backend_ops (llama/10542) Sergio López 2024-11-28 12:51:38 +0100
  • 90dd5fca9c CANN: Fix SOC_TYPE compile bug (llama/10519) leo-pony 2024-11-28 15:25:24 +0800
  • 2490f2a7f8 CANN: ROPE operator optimization (llama/10540) Chenguang Li 2024-11-28 14:24:46 +0800
  • 230e985633 Add some minimal optimizations for CDNA (llama/10498) uvos 2024-11-27 17:10:08 +0100
  • ae24083f23 metal : fix group_norm support condition (llama/0) Georgi Gerganov 2024-11-27 11:22:14 +0200
  • 6463e36369 vulkan: define all quant data structures in types.comp (llama/10440) Jeff Bolz 2024-11-27 01:32:54 -0600
  • b3301f7d82 vulkan: Handle GPUs with less shared memory (llama/10468) Jeff Bolz 2024-11-27 01:30:27 -0600
  • ab5d4d93ec vulkan: further optimize q5_k mul_mat_vec (llama/10479) Jeff Bolz 2024-11-27 01:21:59 -0600
  • 2d6e9dd723 vulkan: skip integer div/mod in get_offsets for batch_idx==0 (llama/10506) Jeff Bolz 2024-11-27 01:08:54 -0600
  • 2f16e51553 vulkan: optimize Q2_K and Q3_K mul_mat_vec (llama/10459) Jeff Bolz 2024-11-27 01:00:50 -0600
  • 0f0994902f mtgpu: Add MUSA_DOCKER_ARCH in Dockerfiles && update cmake and make (llama/10516) R0CKSTAR 2024-11-27 00:00:41 +0800
  • 5e1fcc1780 vulkan: fix group_norm (llama/10496) Jeff Bolz 2024-11-26 09:45:05 -0600
  • 48f421de23 cmake : enable warnings in llama (llama/10474) Georgi Gerganov 2024-11-26 14:18:08 +0200
  • e7afb2b991 ggml-cpu: cmake add arm64 cpu feature check for macos (llama/10487) Charles Xu 2024-11-26 12:37:05 +0100