Commit graph

  • 647abea814
    ggml: new optimization interface (ggml/988) Johannes Gäßler 2024-11-16 22:17:59 +02:00
  • 1333449f89
    scripts : update sync Georgi Gerganov 2024-11-16 22:16:04 +02:00
  • 7853400433 llama/ex: remove --logdir argument Johannes Gäßler 2024-11-16 17:36:23 +01:00
  • 19d012e78a
    ggml : adapt AMX to tensor->grad removal (#0) Georgi Gerganov 2024-11-16 21:38:01 +02:00
  • de51c0220a
    make : add ggml-opt (#0) Georgi Gerganov 2024-11-16 21:35:31 +02:00
  • b235157935
    tests : remove test-grad0 Georgi Gerganov 2024-11-16 21:34:03 +02:00
  • a268e961c8
    ggml : fix compile warnings (#0) Georgi Gerganov 2024-11-16 21:32:41 +02:00
  • e7978da5ee
    ggml: new optimization interface (ggml/988) Johannes Gäßler 2024-11-16 21:31:25 +02:00
  • db4cfd5dbc llamafile : fix include path (#0) b4102 Georgi Gerganov 2024-11-16 17:58:56 +02:00
  • 8ee0d09ae6 make : auto-determine dependencies (#0) Georgi Gerganov 2024-11-16 17:58:32 +02:00
  • 013785029b
    llamafile : fix include path (#0) Georgi Gerganov 2024-11-16 17:58:56 +02:00
  • e13371a163
    make : auto-determine dependencies (#0) Georgi Gerganov 2024-11-16 17:58:32 +02:00
  • bfcff24714
    Merge 1716e6b25a into bcdb7a2386 Xuan Son Nguyen 2024-11-16 18:00:16 +01:00
  • bcdb7a2386
    server: (web UI) Add samplers sequence customization (#10255) b4100 MaggotHATE 2024-11-16 18:26:54 +05:00
  • f7b0233eca
    wip gg/logits-slowdown Georgi Gerganov 2024-11-16 10:04:49 +02:00
  • f245cc28d4
    scripts : fix missing key in compare-llama-bench.py (#10332) Georgi Gerganov 2024-11-16 10:32:50 +02:00
  • 2ba5a9e9fb
    scripts : fix missing key in compare-llama-bench.py Georgi Gerganov 2024-11-16 10:17:34 +02:00
  • 6beb50dbf6
    vulkan: the index in ggml_vk_host_free could be uinitialized if pinned_memory.size() is zero FirstTimeEZ 2024-11-16 20:24:46 +13:00
  • 772703c8ff
    vulkan: Optimize some mat-vec mul quant shaders (#10296) b4098 Jeff Bolz 2024-11-16 00:26:57 -06:00
  • aa51d1dd43
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-16 19:08:13 +13:00
  • 04a071108e
    Merge branch 'ggerganov:master' into server-chat-templates MaggotHATE 2024-11-16 09:30:28 +05:00
  • d283dbcd37 vulkan: change an assertion FirstTimeEZ 2024-11-16 17:17:10 +13:00
  • 1840df1b58 Apply suggestions from the PR: employ CPU buffers to copy results, use correct ctx_size and add GGML_ASSERT to check v_output Lucas Nogueira 2024-11-16 01:13:46 -03:00
  • 25380ef211 Update .clang-format Eric Curtin 2024-11-15 13:47:48 +00:00
  • 3c3543ab56 64 thread experiment Eve 2024-11-15 22:16:02 -05:00
  • dd3a6ce9f8
    vulkan : add cmake preset debug/release (#10306) FirstTimeEZ 2024-11-16 14:59:33 +13:00
  • 1e58ee1318
    ggml : optimize Q4_0 into Q4_0_X_Y repack (#10324) b4096 Dan Johansson 2024-11-16 01:53:37 +01:00
  • 89e4caaaf0
    llama : save number of parameters and the size in llama_model (#10286) b4095 FirstTimeEZ 2024-11-16 13:42:13 +13:00
  • a79d81daa7 docs: vulkan build instructions to use git bash mingw64 FirstTimeEZ 2024-11-16 13:39:12 +13:00
  • 1cb0e11529 Merge branch 'patch-2' of https://github.com/FirstTimeEZ/llama.cpp into patch-2 FirstTimeEZ 2024-11-16 13:33:47 +13:00
  • 15176afd7d llama_model_size is explicity sized as uint64_t FirstTimeEZ 2024-11-16 13:33:40 +13:00
  • 740bbe94c0
    Merge branch 'ggerganov:master' into vulkan-cmake-preset FirstTimeEZ 2024-11-16 12:16:20 +13:00
  • 932f28e261
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-16 12:16:05 +13:00
  • a3dc73df7a
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-16 12:15:58 +13:00
  • 8007cb0115 ggml: Optimize Q4_0 into Q4_0_X_Y repack Dan Johansson 2024-11-13 13:47:59 +01:00
  • 74d73dc85c
    Make updates to fix issues with clang-cl builds while using AVX512 flags (#10314) b4094 Srihari-mcw 2024-11-16 02:57:00 +05:30
  • 3b837fea7e rename "name" --> "label" Xuan Son Nguyen 2024-11-15 22:26:58 +01:00
  • 4047be74da
    scripts: update compare-llama-bench.py (#10319) b4093 Johannes Gäßler 2024-11-15 21:19:03 +01:00
  • 883d206fbd ggml : fix some build issues b4092 slaren 2024-11-15 20:20:54 +01:00
  • d976654e50 CUDA: remove DMMV, consolidate F16 mult mat vec Johannes Gäßler 2024-11-12 20:49:35 +01:00
  • 12d5491db9 ggml : fix some build issues slaren 2024-11-15 20:20:54 +01:00
  • 4ee5d520ce Update spacing Srihari-mcw 2024-11-15 23:21:49 +05:30
  • 5c1d1177d3 Add complete implementation of the classical PCA algorithm with covariance matrix and power iteration with a very simple test file Lucas Nogueira 2024-11-15 14:32:27 -03:00
  • 7f9e2f16a4 Modify and use settings-modal-short-input MaggotHATE 2024-11-15 21:34:58 +05:00
  • 6e9d976fe7
    vulkan: build instructions to use git bash mingw64 FirstTimeEZ 2024-11-16 04:10:08 +13:00
  • 7ab4f08669
    vulkan: build instructions to use git bash mingw64 FirstTimeEZ 2024-11-16 04:00:26 +13:00
  • 71b47841d5 scripts: update compare-llama-bench.py Johannes Gäßler 2024-11-15 15:25:45 +01:00
  • a9114b358a
    Merge branch 'ggerganov:master' into vulkan-cmake-preset FirstTimeEZ 2024-11-16 03:17:47 +13:00
  • 4e856e30ef
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-16 03:17:39 +13:00
  • 224f70a6d7
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-16 03:17:30 +13:00
  • 09ecbcb596 cmake : fix ppc64 check (whisper/0) b4091 Georgi Gerganov 2024-11-15 15:35:22 +02:00
  • 3225008973 ggml : vulkan logs (whisper/2547) thewh1teagle 2024-11-15 15:33:53 +02:00
  • cbf5541a82 sync : ggml Georgi Gerganov 2024-11-15 15:31:16 +02:00
  • 3f21ccf38b
    cmake : fix ppc64 check (whisper/0) Georgi Gerganov 2024-11-15 15:35:22 +02:00
  • 61063f42c7
    ggml : vulkan logs (whisper/2547) thewh1teagle 2024-11-15 15:33:53 +02:00
  • ff721fefee
    sync : ggml Georgi Gerganov 2024-11-15 15:31:16 +02:00
  • 980d20d117 Make updates to fix issues with clang-cl builds while using AVX512 flags Srihari-mcw 2024-11-15 05:05:31 -08:00
  • 7a24341e1b
    Merge branch 'ggerganov:master' into server-chat-templates MaggotHATE 2024-11-15 18:01:38 +05:00
  • 68ca77a148 chore : Fix the error when compiling rocm build on windows using cmake (#9666) cocochick 2024-11-15 20:37:38 +08:00
  • 5d48927370 Add .clang-format file Eric Curtin 2024-11-15 12:12:50 +00:00
  • 18429220bd
    AVX BF16 and single scale quant optimizations (#10212) b4088 Eve 2024-11-15 11:47:58 +00:00
  • f0204a0ec7
    ci: build test musa with cmake (#10298) b4087 R0CKSTAR 2024-11-15 19:47:25 +08:00
  • 7e2226a27d
    Merge branch 'ggerganov:master' into vulkan-cmake-preset FirstTimeEZ 2024-11-16 00:22:48 +13:00
  • c2410e2b05
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-16 00:22:38 +13:00
  • eb731c0012
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-16 00:22:29 +13:00
  • 57f8355b29
    sycl: Update Intel docker images to use DPC++ 2025.0 (#10305) Romain Biessy 2024-11-15 12:10:45 +01:00
  • d3bddeacbf
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-15 23:23:49 +13:00
  • fcf90a1d23
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-15 23:23:38 +13:00
  • cb6a963b8d
    Merge branch 'ggerganov:master' into vulkan-cmake-preset FirstTimeEZ 2024-11-15 23:23:26 +13:00
  • 1333a47641 Merge branch 'master' into romain/set_arch romain.biessy 2024-11-15 10:22:59 +00:00
  • 35b2287f26 vulkan: cmake preset debug/release FirstTimeEZ 2024-11-15 23:04:26 +13:00
  • ac3c3cc9ec sycl: Update Intel docker images to use DPC++ 2025.0 romain.biessy 2024-11-15 09:56:18 +00:00
  • 9901068ac7
    server : (web UI) add copy button for code block, fix api key (#10242) b4085 Xuan Son Nguyen 2024-11-15 05:48:49 -04:00
  • 469bb57456
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-15 20:15:58 +13:00
  • 2d4eab4e5f
    Merge branch 'ggerganov:master' into patch-4 FirstTimeEZ 2024-11-15 20:15:45 +13:00
  • 231f9360d9
    cann: dockerfile and doc adjustment (#10302) Chenguang Li 2024-11-15 15:09:35 +08:00
  • 4714ec3fa8
    vulkan install instructions git bash mingw64 FirstTimeEZ 2024-11-15 19:47:58 +13:00
  • 4802ad350b
    scripts : fix regex in sync [no ci] Georgi Gerganov 2024-11-15 08:38:43 +02:00
  • b13f7e802b cann: dockerfile and doc adjustment noemotiovon 2024-11-15 06:27:24 +00:00
  • 5718cbb2f7
    Merge branch 'ggerganov:master' into server-chat-templates MaggotHATE 2024-11-15 09:28:51 +05:00
  • e9b707ebee
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-15 16:47:21 +13:00
  • 5a54af4d4f
    sycl: Use syclcompat::dp4a (#10267) b4082 Romain Biessy 2024-11-15 04:09:12 +01:00
  • c7b8ab73de vulkan: Optimize soft_max Jeff Bolz 2024-11-14 20:18:57 -06:00
  • fe792d62b1
    Merge pull request #26 from NexaAI/master Zack Li 2024-11-14 18:16:50 -08:00
  • 25190fefa2
    Merge pull request #25 from NexaAI/weili/master-release Zack Li 2024-11-14 17:49:34 -08:00
  • b1de101f27 ci: build test musa with cmake Xiaodong Ye 2024-11-15 09:16:24 +08:00
  • f281ca3b53 merge Eve 2024-11-14 19:59:07 -05:00
  • 5d3e7c3579
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-15 13:47:17 +13:00
  • e4ca946c48 free omni_ctx heap malloc space in omni_free() api Currently mem leaks in qwen2audio are almost fixed. 李为 2024-11-15 08:31:01 +08:00
  • 1607a5e5b0
    backend cpu: add online flow for aarch64 Q4_0 GEMV/GEMM kernels (#9921) b4081 Charles Xu 2024-11-15 01:28:50 +01:00
  • 9352321b49 pull in master (same changes removed) Eve 2024-11-14 19:18:56 -05:00
  • e343f74167
    Merge branch 'ggerganov:master' into patch-2 FirstTimeEZ 2024-11-15 12:20:17 +13:00
  • 749a9e5e66 Merge remote-tracking branch 'origin/master' into feature/online-flow slaren 2024-11-14 22:40:59 +01:00
  • a97a52bb02 vulkan: Optimize some mat-vec mul quant shaders Jeff Bolz 2024-11-13 12:38:32 -06:00
  • ae8de6d50a
    ggml : build backends as libraries (#10256) b4080 Diego Devesa 2024-11-14 18:04:35 +01:00
  • 56b7f93aef
    Merge c5d8bb5a81 into 4a8ccb37ad Meng, Hengyu 2024-11-14 11:00:25 -05:00
  • dc228ddaa3
    Merge cd457dce20 into 4a8ccb37ad Jhen-Jie Hong 2024-11-14 11:00:25 -05:00
  • 33e6df58ba
    Merge 66af1a4b4c into 4a8ccb37ad Felix Yun 2024-11-14 11:00:25 -05:00
  • f2f5c3b63d
    ggml: separate musa into its own section in the Makefile (#10294) R0CKSTAR 2024-11-14 23:24:35 +08:00
  • 5430726711 Rename GGML_SYCL_ARCH to GGML_SYCL_DEVICE_ARCH romain.biessy 2024-11-14 15:22:13 +00:00