Commit Graph

  • e81c67a125 bench : start benching the decoder Georgi Gerganov 2023-09-14 10:06:14 +03:00
  • f408c64564 bench : fix timings by running a pre-heat Georgi Gerganov 2023-09-13 23:02:33 +03:00
  • d863f725a1 coreml : add code to toggle Core ML config (CPU, ANE, GPU) Georgi Gerganov 2023-09-13 22:51:10 +03:00
  • d37f56e7a9 ios : update submodule Georgi Gerganov 2023-09-13 21:31:29 +03:00
  • 23277d21ce readme : add Metal info Georgi Gerganov 2023-09-13 20:54:03 +03:00
  • ecb23fb1eb metal : sync latest llama.cpp kernels Georgi Gerganov 2023-09-13 20:44:05 +03:00
  • 8e8daa8451 metal : speed-up KQ multiplication Georgi Gerganov 2023-09-13 19:59:16 +03:00
  • 16db4da3f1 swiftui : fix build Georgi Gerganov 2023-09-13 19:49:11 +03:00
  • 257d7942af ios : add Metal support Georgi Gerganov 2023-09-13 19:45:12 +03:00
  • 181bb8cb28 objc : fix build (no Metal yet) Georgi Gerganov 2023-09-13 18:54:41 +03:00
  • 796f84cd95 whisper : add <functional> header Georgi Gerganov 2023-09-13 13:35:42 +03:00
  • 77f4bf49c8 cmake : update to support Metal build Georgi Gerganov 2023-09-13 13:34:51 +03:00
  • b6f09669a2 whisper : factor out alloc init in a function Georgi Gerganov 2023-09-13 12:51:52 +03:00
  • 254b687239 whisper : add whisper_allocr to wrap ggml_allocr Georgi Gerganov 2023-09-13 11:58:19 +03:00
  • b19888cfb4 ggml-alloc : try to make CI happy by reducing vram to 128GB Georgi Gerganov 2023-09-13 11:57:46 +03:00
  • 905c944143 ggml : use simpler ggml_bytes() implementation Georgi Gerganov 2023-09-13 11:39:09 +03:00
  • 3074a7ff14 whisper : offload the Encoder to Metal Georgi Gerganov 2023-09-13 00:09:44 +03:00
  • ec9a7db74c whisper : remove ggml_repeat in the encoder Georgi Gerganov 2023-09-12 20:34:32 +03:00
  • cd476375b4 metal : run "cross" step on the GPU Georgi Gerganov 2023-09-12 20:11:13 +03:00
  • 9fdd415367 ggml : fix ggml_nbytes (probably temp solution) Georgi Gerganov 2023-09-12 20:10:53 +03:00
  • 79a88057bd metal : add multi-decoder support Georgi Gerganov 2023-09-12 19:33:29 +03:00
  • fbc9ddc582 metal : decoder works on GPU! Georgi Gerganov 2023-09-12 19:23:30 +03:00
  • 3b9979a373 ci : try to debug vmem issue Georgi Gerganov 2023-09-12 14:08:48 +03:00
  • de94c783ee Merge branch 'master' into metal-and-alloc Georgi Gerganov 2023-09-12 14:02:43 +03:00
  • 3fec2119e6 whisper : fix bench regression + fix performance when using CPU BLAS (#1275) Georgi Gerganov 2023-09-12 13:54:04 +03:00
  • 09a6325de5 ggml : use sched_yield when using BLAS + add comment fix-bench Georgi Gerganov 2023-09-12 13:33:09 +03:00
  • 39c4fc59dd whisper : fix bench regression Georgi Gerganov 2023-09-12 11:21:02 +03:00
  • d3b2dd4955 whisper : initial Metal version Georgi Gerganov 2023-09-11 15:37:24 +03:00
  • 4845b9ed09 whisper.android : try to fix build Georgi Gerganov 2023-09-11 15:19:21 +03:00
  • 2770d46ef5 whisper : refactor ggml-alloc init Georgi Gerganov 2023-09-11 15:04:33 +03:00
  • 4d9acc60c3 ci : see if this is causing the crash Georgi Gerganov 2023-09-10 23:53:49 +03:00
  • 06d1d2836b extra : update sync-ggml.sh script to also sync ggml-alloc Georgi Gerganov 2023-09-10 22:45:38 +03:00
  • 9a78b72246 ios : update submodule Georgi Gerganov 2023-09-10 22:36:50 +03:00
  • 794e8fe0ea build : fix ggml-alloc Georgi Gerganov 2023-09-10 22:19:39 +03:00
  • fa672b46e6 whisper : CoreML support ggml-alloc Georgi Gerganov 2023-09-10 21:55:20 +03:00
  • af6f67b251 whisper : ggml-alloc is now supported Georgi Gerganov 2023-09-10 20:09:17 +03:00
  • bed5ad69dd whisper : allocate encoder and decoder using ggml-alloc Georgi Gerganov 2023-09-10 19:50:34 +03:00
  • 949ab6328d whisper : factor out graph builds Georgi Gerganov 2023-09-10 19:23:06 +03:00
  • fbc3f8033e metal : init Georgi Gerganov 2023-09-10 18:38:34 +03:00
  • 9b14418863 whisper : faster beam_search sampling via reduced KV cache copies (#1243) bobqianic 2023-09-10 21:04:27 +08:00
  • 3c50be2217 whisper : remove comment grammar-debug Georgi Gerganov 2023-09-10 13:27:06 +03:00
  • 37de5dcf2b command : enable beam-search, add "no_timestamps", add "context", add p Georgi Gerganov 2023-09-10 12:22:57 +03:00
  • 7a2abb311d grammars : add assistant + update comments Georgi Gerganov 2023-09-09 20:24:58 +03:00
  • 54d168db67 command : grammar-related improvements Georgi Gerganov 2023-09-09 17:58:54 +03:00
  • 6ddc727fac java : fixed signing of java artifact using gradle (#1267) Nicholas Albion 2023-09-10 01:55:51 +10:00
  • 9c1a414feb use secrets.GPG_PRIVATE_KEY and GPG_PASSPHRASE feature/debug-gradle-signing Nicholas Albion 2023-09-10 00:49:13 +10:00
  • a992af5746 use GPG_PRIVATE_KEY and GPG_PASSPHRASE Nicholas Albion 2023-09-10 00:17:51 +10:00
  • 1840090261 cd bindings/java Nicholas Albion 2023-09-09 14:37:36 +10:00
  • 3c35fc0c5c added temporary step "Debug gradle signing" Nicholas Albion 2023-09-09 13:59:29 +10:00
  • 0badc1483d --stacktrace signMavenJavaPublication Nicholas Albion 2023-09-09 13:39:01 +10:00
  • acb5278cc8 ci : try to fix gradle action (#1265) Georgi Gerganov 2023-09-08 20:50:15 +03:00
  • 0839209cab gitignore : update Georgi Gerganov 2023-09-08 19:45:28 +03:00
  • b39809668a sync : ggml (HBM + Metal + style) (#1264) Georgi Gerganov 2023-09-08 17:58:31 +03:00
  • 3e9edc6845 ci : upgrade gradle to 2.4.2 (#1263) Georgi Gerganov 2023-09-08 17:58:14 +03:00
  • bfc73f1fa2 sync : ggml (CUDA faster rope) Georgi Gerganov 2023-09-08 15:01:26 +03:00
  • f00c9bba33 cmake : noramlize case (#1129) Georgi Gerganov 2023-09-08 14:50:03 +03:00
  • b55b505690 build : do not use _GNU_SOURCE gratuitously (#1129) Przemysław Pawełczyk 2023-09-07 11:36:14 +02:00
  • 2818de21ff examples : fix build + compile warnings (close #1256) Georgi Gerganov 2023-09-07 12:33:12 +03:00
  • aed5d40607 models : add quantum models to download-ggml-model.sh (#1235) Neil Chudleigh 2023-09-07 05:16:58 -04:00
  • afa5477d1c whisper.android : bump gradle plugin and dependencies + a lint pass (#1255) Digipom 2023-09-07 05:15:59 -04:00
  • 01fcd42431 sign jar for Maven Central repo Nicholas Albion 2023-09-07 11:45:01 +10:00
  • f990610776 whisper.android : address ARM's big.LITTLE arch by checking cpu info (#1254) Digipom 2023-09-06 11:32:30 -04:00
  • 64cb45fd79 make : fix detection of AVX2 on macOS (#1250) Didzis Gosko 2023-09-06 18:22:21 +03:00
  • ace6c12ec6 ggml : posixify pagesize (#1251) Przemysław Pawełczyk 2023-09-06 17:19:36 +02:00
  • b8f34d1ed7 whisper : fine-tuning grammar functionality Georgi Gerganov 2023-09-06 17:05:05 +03:00
  • 97ebb48b99 command : fix exception when recognizing the command Georgi Gerganov 2023-09-06 15:05:17 +03:00
  • b0306cd5cf build : fix after master merge Georgi Gerganov 2023-09-06 14:27:51 +03:00
  • afc84b35b0 Merge branch 'master' into HEAD Georgi Gerganov 2023-09-06 13:33:34 +03:00
  • cac75be05b configured publishing.repositories Nicholas Albion 2023-09-06 13:13:36 +10:00
  • c3f319d7c2 ggml : sync latest llama.cpp (view_src + alloc improvements) (#1247) Georgi Gerganov 2023-09-05 20:57:27 +03:00
  • ba3c333611 make : improve cpuinfo handling on x86 hosts (#1238) Przemysław Pawełczyk 2023-09-05 13:58:47 +02:00
  • 59a3d0cb57 ggml : sync (ggml-alloc, GPU, eps, etc.) (#1220) Georgi Gerganov 2023-09-05 13:54:40 +03:00
  • 6780c98e19 readme : update CMake build commands (#1231) 布客飞龙 2023-09-05 18:53:34 +08:00
  • 2f52783a08 OSSRH_USERNAME -> JIRA_USER Nicholas Albion 2023-08-31 14:54:02 +10:00
  • 7dec9d8cc4 build-root-directory: bindings/java Nicholas Albion 2023-08-31 12:04:16 +10:00
  • 31476ccc0e whisper : add grammar-based sampling Evan Jones 2023-08-29 22:56:29 -04:00
  • fb0a24fba2 ci : enable java package publishing (#1228) Georgi Gerganov 2023-08-31 02:57:43 +03:00
  • 3efe146d27 ci : enable java package publishing java-bindings Georgi Gerganov 2023-08-30 22:13:38 +03:00
  • 8e30bf3c02 ggml : fix compilation errors incurred by -Werror (#1227) ChangSeok Oh 2023-08-30 12:09:15 -07:00
  • 99d3c105f5 whisper.android : fix cmake multiple libraries build (#1224) Jhen-Jie Hong 2023-08-30 19:45:13 +08:00
  • 18e9889418 coreml : wrap inference call in @autoreleasepool to fix memory leak (#1218) Dener Stassun 2023-08-29 09:44:38 -03:00
  • 8e46ba80d3 make : use cpuinfo in MSYS2 to enable x86 ISA extensions on the host (#1216) Przemysław Pawełczyk 2023-08-28 12:28:26 +02:00
  • b0d35995c4 make : add support for building on DragonFlyBSD/NetBSD/OpenBSD (#1212) Przemysław Pawełczyk 2023-08-27 20:38:46 +02:00
  • 25466aa1c3 ggml : fix compiling when SSE3 is available but not SSSE3 (#1210) Przemysław Pawełczyk 2023-08-27 20:37:31 +02:00
  • 601c2d2181 ggml : detect SSSE3 (#1211) Przemysław Pawełczyk 2023-08-27 20:36:41 +02:00
  • 175ffa64ee examples : vim plugin and LSP server (#1144) AustinMroz 2023-08-27 13:35:06 -05:00
  • cb5fb0a12d whisper : initial hipBLAS support (#1209) ardfork 2023-08-27 17:03:58 +00:00
  • b5bb5c85d4 whisper : allow whisper_full from mel spectrogram - no audio (#1214) Georgi Gerganov 2023-08-27 20:02:57 +03:00
  • 7e54df414e whisper : significantly improve the inference quality (#1148) bobqianic 2023-08-28 00:51:33 +08:00
  • 20a80972f4 whisper.android : migrate from ndk-build to CMake (#1204) junkfood 2023-08-28 00:35:16 +08:00
  • 7ef3f3837e main : log probs to text file (#1205) Yunès 2023-08-27 18:09:06 +02:00
  • aad2dad38a whisper : minor fixes (#1154) Fangjun Kuang 2023-08-28 00:02:00 +08:00
  • 66f2078878 build : fix OpenBLAS detection under Arch Linux (#1173) Marcin Mielniczuk 2023-08-25 18:26:34 +02:00
  • 8ce20f0f3d make : fix Linux machines supporting AVX1 not AVX2 (#1162) Eric Swanson 2023-08-25 08:52:22 -04:00
  • c84cf87261 whisper : add precalculated values of sin/cos for speeding up FFT (#1142) Alexandr Graschenkov 2023-08-25 16:51:14 +04:00
  • c5f9acf4b7 make : simplify Makefile (#1147) alonfaraj 2023-08-25 15:20:44 +03:00
  • 7decc85eb7 cmake : fix PowerPC build failures introduced in #1174 (#1196) Marcin Mielniczuk 2023-08-25 14:19:48 +02:00
  • 21e8c67a4f Fix AVX etc. under GCC/CMake (#1174) Marcin Mielniczuk 2023-08-19 20:39:03 +02:00
  • a4bb2df36a quantize : fix load vocab crash when len is 128 (#1160) Jhen-Jie Hong 2023-08-06 16:04:42 +08:00
  • b948361956 examples : add tinydiarization support for streaming (#1137) Duncan McConnell 2023-08-03 03:24:07 -05:00