wylab/llama.cpp
1
0
Fork 1
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-16 09:25:56 +02:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • e7da30b584 fix: Viewing multiple PDF attachments (#16974) Aleksander Grygier 2025-11-03 18:53:26 +01:00
  • ed8aa63320 model-conversion : pass config to from_pretrained (#16963) Daniel Bevenius 2025-11-03 18:01:59 +01:00
  • 48bd26501b server : add props.model_alias (#16943) b6937 Georgi Gerganov 2025-11-03 15:38:23 +02:00
  • 622cd010ff ggml: CUDA: add head size 72 for flash-attn (#16962) b6936 theo77186 2025-11-03 14:29:11 +01:00
  • 070ff4d535 mtmd: add --image-min/max-tokens (#16921) b6935 Xuan-Son Nguyen 2025-11-03 11:11:18 +01:00
  • bf7b0c9725 mtmd: pad mask for qwen2.5vl (#16954) b6934 Xuan-Son Nguyen 2025-11-03 10:25:55 +01:00
  • fcfce040e8 ggml : LoongArch fixes (#16958) b6933 Jinyang He 2025-11-03 14:40:02 +08:00
  • ee3a5a10ad sync: minja (glm 4.6 & minmax m2 templates) (#16949) b6932 Olivier Chafik 2025-11-03 05:33:56 +00:00
  • 7e994168b1 SYCL: optimized repeat_back kernel (3× fewer asm instructions, 2× faster)Feature/sycl repeat back opt (#16869) b6931 shani-f 2025-11-03 03:35:33 +02:00
  • bcfa87622a feat(webui): improve LaTeX rendering with currency detection (#16508) Sascha Rogmann 2025-11-03 00:41:08 +01:00
  • a2054e3a8f test-backend-ops : fix segfault in moe-expert-reduce test in support mode and coverage (#16936) b6929 Shagun Bera 2025-11-03 04:40:30 +05:30
  • dd52868050 ci : disable failing riscv cross build (#16952) Sigbjørn Skjæret 2025-11-02 23:11:21 +01:00
  • 6b9a52422b model: add Janus Pro for image understanding (#16906) b6927 Zhiyong Wang 2025-11-02 13:08:04 -08:00
  • 2f966b8ed8 clip : use FA (#16837) Georgi Gerganov 2025-11-02 22:21:48 +02:00
  • d441c31b19 metal : remove stray return gg/clip-fa Georgi Gerganov 2025-11-02 18:24:00 +02:00
  • cd5e3b5754 server : support unified cache across slots (#16736) Georgi Gerganov 2025-11-02 18:14:04 +02:00
  • 87c9efc3b2 common : move gpt-oss reasoning processing to init params (#16937) b6924 Aldehir Rojas 2025-11-02 08:56:28 -06:00
  • cdb3deae76 trailing space Xuan Son Nguyen 2025-11-02 12:12:09 +01:00
  • b67a168f10 improve debugging message Xuan Son Nguyen 2025-11-02 12:08:49 +01:00
  • 76af40aaaa docs: remove llama_sampler_accept reference in sampling sample usage (#16920) b6923 Adrian Lundberg 2025-11-02 10:28:37 +01:00
  • 29330dcb55 cont : remove obsolete comment [no ci] Georgi Gerganov 2025-11-02 10:16:17 +02:00
  • bdb43f6e9c clip : print more detailed op support info during warmup Georgi Gerganov 2025-11-02 10:13:48 +02:00
  • 7db35a7958 CUDA: add FLOOR, CEIL, ROUND, TRUNC unary ops (#16917) b6922 mnehete32 2025-11-02 08:42:57 +05:30
  • a864132ba5 devops: fix failing s390x docker build (#16918) Aaron Teo 2025-11-02 08:48:46 +08:00
  • d38d9f0877 ggml: add s390x cpu-feats (#16774) b6920 Aaron Teo 2025-11-02 08:48:23 +08:00
  • b4955f0ae6 implement "auto" mode for clip flash attn Xuan Son Nguyen 2025-11-01 23:52:40 +01:00
  • 7fd205a8e8 scripts : add script to bench models (#16894) b6919 Georgi Gerganov 2025-11-02 00:15:31 +02:00
  • 19116a4b38 Merge branch 'master' into gg/clip-fa Xuan Son Nguyen 2025-11-01 23:08:56 +01:00
  • 2f68ce7cfd webui: auto-refresh /props on inference start to resync model metadata (#16784) Pascal 2025-11-01 19:49:51 +01:00
  • e4a71599e5 webui: add HTML/JS preview support to MarkdownContent with sandboxed iframe (#16757) Pascal 2025-11-01 17:14:54 +01:00
  • dd5e8cab51 vendor : update cpp-httplib to 0.27.0 (#16846) b6916 Adrien Gallouët 2025-11-01 16:52:17 +01:00
  • cf659bbb8e mtmd: refactor preprocessing + support max/min pixels (#16878) b6915 Xuan-Son Nguyen 2025-11-01 15:51:36 +01:00
  • d8b860a219 Add a setting to display message generation statistics (#16901) Aleksander Grygier 2025-11-01 15:35:57 +01:00
  • 1ae74882f8 webui: recognize AsciiDoc files as valid text files (#16850) Jaromír Hradílek 2025-11-01 15:02:57 +01:00
  • 961660b8c3 common : allow --system-prompt-file for diffusion-cli (#16903) b6912 Sigbjørn Skjæret 2025-11-01 11:01:42 +01:00
  • 74fef4129f codeowners : update after refactor (#16905) Sigbjørn Skjæret 2025-11-01 08:55:25 +01:00
  • 5d8bb900bc vulkan: Fix multi_add invalid descriptor usage (#16899) b6910 Jeff Bolz 2025-11-01 00:52:14 -05:00
  • 2e76e01360 vulkan: fuse mul_mat+add and mul_mat_id+add_id (#16868) b6909 Jeff Bolz 2025-11-01 00:45:28 -05:00
  • d3dc9dd898 CUDA: Remove unneded bias/gate dims in fused mmvq (#16858) b6908 Oliver Simons 2025-11-01 06:13:26 +01:00
  • bea04522ff refactor : llama-model.cpp (#16252) b6907 Piotr Wilkin (ilintar) 2025-10-31 23:40:23 +01:00
  • 0de0a01576 model : Minimax M2 (#16831) b6906 Piotr Wilkin (ilintar) 2025-10-31 21:20:47 +01:00
  • e58d585604 model : add Granite Hybrid nano types (#16896) b6905 Giuseppe Scrivano 2025-10-31 21:20:07 +01:00
  • 31c511a968 CUDA: Volta tensor core support for MMF (#16843) b6904 Johannes Gäßler 2025-10-31 15:57:19 +01:00
  • 6d39015a74 sync : ggml Georgi Gerganov 2025-10-31 16:25:50 +02:00
  • 4146d6a1a6 CUDA: add expert reduce kernel (#16857) Aman Gupta 2025-10-31 20:05:07 +08:00
  • 8da3c0e200 batch : fix consistency checks for the input positions (#16890) b6901 Georgi Gerganov 2025-10-31 13:50:33 +02:00
  • c22473b580 server : don't print user inputs to console (#16871) b6900 Georgi Gerganov 2025-10-31 10:54:19 +02:00
  • 0f715b4e75 server : fix typos in server.cpp comments [no ci] (#16883) Daniel Bevenius 2025-10-31 09:51:26 +01:00
  • d2d931f173 vulkan: disable spirv-opt for rope shaders (#16872) b6898 Jeff Bolz 2025-10-31 02:34:47 -05:00
  • 2976b0374d vulkan: Fix crash when FP16 mul_mat accumulation is not supported (#16796) b6897 Masato Nakasaka 2025-10-31 16:18:59 +09:00
  • d2a2673dd1 vulkan: fix shmem overrun in mmq id shader (#16873) b6896 Ruben Ortlam 2025-10-31 08:14:49 +01:00
  • 13002a0896 ggml-hexagon: respect input size when getting/setting tensor data (#16836) b6895 l3utterfly 2025-10-31 12:46:31 +08:00
  • 6eb208d17e ci : enable free-disk-space on cuda docker build (#16877) b6894 Sigbjørn Skjæret 2025-10-31 00:34:27 +01:00
  • 9984cbb61d opencl: fix boundary handling for mul_mm (#16875) lhez 2025-10-30 16:00:20 -07:00
  • ce18efeaf1 convert : update transformers requirements (#16866) RodriMora 2025-10-30 23:15:03 +01:00
  • 16724b5b68 server : bump request URI max length to 32768 (#16862) b6891 chansikpark 2025-10-30 14:22:23 -04:00
  • b52edd2558 server : remove n_past (#16818) b6890 Georgi Gerganov 2025-10-30 18:42:57 +02:00
  • a4b54f2697 cont : add warning about unsupported ops Georgi Gerganov 2025-10-30 18:15:34 +02:00
  • 3aa835bfe6 clip : use FA Georgi Gerganov 2025-10-29 11:28:45 +02:00
  • 517b7170e1 cpu: introduce chunking for repack matmuls and enable matmul-id chunking on ARM64 (#16833) b6889 Max Krasnyansky 2025-10-30 09:06:13 -07:00
  • 835e918d84 common: fix typo in cli help text (#16864) b6888 Shagun Bera 2025-10-30 21:17:31 +05:30
  • d261223d24 model: add support for qwen3vl series (#16780) b6887 JJJYmmm 2025-10-30 23:19:14 +08:00
  • dcca0d3ab8 cpu: introduce chunking for flash attention (#16829) b6886 Max Krasnyansky 2025-10-30 05:26:05 -07:00
  • bacddc049a model: Add support for CogVLM model (#15002) b6885 Tianyue-Zhao 2025-10-30 07:18:50 -04:00
  • 229bf68628 cuda : fix argsort with 64k+ rows (#16849) b6884 Sigbjørn Skjæret 2025-10-30 08:56:28 +01:00
  • d7395115ba llama : use std::abs instead of abs (#16853) b6883 Jan Boon 2025-10-30 14:30:58 +08:00
  • 052df28b0e vulkan: Handle argsort with a large number of rows (#16851) b6882 Jeff Bolz 2025-10-30 01:27:41 -05:00
  • 8b11deea46 Hide latency of bias and gate-loading (#16847) b6881 Oliver Simons 2025-10-30 04:34:15 +01:00
  • b9ce940177 vulkan: Fuse rope+set_rows (#16769) b6880 Jeff Bolz 2025-10-29 15:13:10 -05:00
  • 3464bdac37 llama: fix ASAN error with M-RoPE (#16848) b6879 Xuan-Son Nguyen 2025-10-29 20:11:39 +01:00
  • e3af5563bd llama: store mrope data in KV cell (#16825) b6878 Xuan-Son Nguyen 2025-10-29 18:09:18 +01:00
  • 10fcc41290 vulkan: Update topk_moe fusion to handle gpt's late softmax (#16656) b6877 Jeff Bolz 2025-10-29 08:44:29 -05:00
  • bcf5bda6f5 Vulkan MMQ Integer Dot Refactor and K-Quant support (#16536) b6876 Ruben Ortlam 2025-10-29 14:39:03 +01:00
  • 3eb2be1ca5 Hexagon Op queue & dispatch optimizations (#16820) b6875 Max Krasnyansky 2025-10-29 06:29:12 -07:00
  • e41bcce8f0 CUDA: use fastdiv in set-rows (#16834) b6874 Aman Gupta 2025-10-29 21:11:53 +08:00
  • 144a4ce824 vendor : sync minja (#16500) b6873 Sigbjørn Skjæret 2025-10-29 14:09:50 +01:00
  • f549b0007d vulkan: Call ggml_vk_buffer_write_2d from ggml_vk_buffer_copy (#16793) b6872 Jeff Bolz 2025-10-29 03:53:04 -05:00
  • 9a3ea685b9 CUDA: Fix bug in topk-moe for gpt-oss (#16821) b6871 Aman Gupta 2025-10-29 15:55:06 +08:00
  • 338074c383 sycl: add RMS_NORM_BACK operation support (#16808) b6870 YaelLogic 2025-10-29 08:14:39 +02:00
  • 851553ea6b cuda: add SET operation support (#16804) b6869 YaelGitAccount 2025-10-28 21:10:28 +02:00
  • 85a7d8677b memory : remove KV cache size padding (#16812) b6868 Georgi Gerganov 2025-10-28 20:19:44 +02:00
  • a8ca18b4b8 llama-bench : clarify benchmarked parts of the computation (#16823) Georgi Gerganov 2025-10-28 19:41:43 +02:00
  • 8284efc35c initialise buffer.device in ggml_hexagon_session (#16816) b6866 l3utterfly 2025-10-28 23:16:20 +08:00
  • 1c1409e131 embedding: add raw option for --embd-output-format (#16541) b6865 Sam Malayek 2025-10-28 03:51:41 -07:00
  • 7a0e900e36 llama: consistent ctx <-> buf order for KV cache (#16746) b6864 Johannes Gäßler 2025-10-28 11:23:54 +01:00
  • 280d97be96 grammar : support array references in json schema (#16792) b6863 Aldehir Rojas 2025-10-28 03:37:52 -05:00
  • 3479efd112 CANN: Improve device ID handling and aclnnArange checks (#16752) b6862 Chenguang Li 2025-10-28 10:54:53 +08:00
  • 463bbf20bf CUDA: add unused vars to mmvf and mmvq (#16807) b6861 Aman Gupta 2025-10-28 10:31:21 +08:00
  • ad8d36beff sycl: add SSM_CONV operation support (#16800) b6860 tamarPal 2025-10-28 03:50:33 +02:00
  • c053e18a66 chat: Add LFM2 tool handling (#16763) b6859 Yuri Khrustalev 2025-10-27 18:54:01 -04:00
  • e1ab084803 mtmd : fix idefics3 preprocessing (#16806) b6858 Xuan-Son Nguyen 2025-10-27 23:12:16 +01:00
  • 5a4ff43e7d llama : disable pipeline parallelism if compute buffer allocation fails (#16748) b6857 Diego Devesa 2025-10-27 13:51:28 -07:00
  • 10640e31aa ggml : fix interpolate with align-corners and ne=1 (#16700) b6856 Acly 2025-10-27 21:50:22 +01:00
  • 80d28f104c HIP: fix AMDGPU_TARGETS, update documentation (#16803) b6855 Johannes Gäßler 2025-10-27 21:39:49 +01:00
  • c55d53acec model : add LightOnOCR-1B model (#16764) b6854 Xuan-Son Nguyen 2025-10-27 16:02:58 +01:00
  • 945501f5ea llama: fix leaked buffers for mmap + split files (#16765) b6853 Johannes Gäßler 2025-10-27 09:17:31 +01:00
  • 75cbdd3fce test-backend-ops: print failed tests at the end (#16785) b6852 Aman Gupta 2025-10-27 09:25:10 +08:00
  • 2b9bd9bf4e sycl: add ROLL operation support (#16665) b6851 tamarPal 2025-10-27 03:20:24 +02:00
  • 59fc1ec8e8 sycl: add REPEAT_BACK operation support (#16734) b6850 shani-f 2025-10-27 03:19:50 +02:00
  • 75d33b9302 CUDA: support for weight clamp in top-k norm (#16702) b6849 Aman Gupta 2025-10-27 09:06:16 +08:00