Commit graph

  • b7a24af083
    Add twinny vscode extension to Extensions and Plugins (#1950) Richard Macarthy 2024-01-31 14:25:06 +00:00
  • c8b1f2369e remove unnecessary parse raw Michael Yang 2024-01-30 16:24:40 -08:00
  • 72b12c3be7 Bump llama.cpp to b1999 Daniel Hiltgen 2024-01-29 12:58:17 -08:00
  • 0632dff3f8
    trim chat prompt based on llm context size (#1963) Bruce MacDonald 2024-01-30 15:59:29 -05:00
  • 509e2dec8a
    Update README.md (#2252) Maximilian Weber 2024-01-30 20:56:51 +01:00
  • 78a48de804
    Merge pull request #2256 from dhiltgen/container_logs Daniel Hiltgen 2024-01-30 08:12:48 -08:00
  • e7dbb00331 Add container hints for troubleshooting Daniel Hiltgen 2024-01-29 08:52:41 -08:00
  • c3f9538636 remove default.nix Marc Raiser 2024-01-29 00:05:07 -05:00
  • 2e06ed01d5 remove unknown CPPFLAGS option Jeffrey Morgan 2024-01-28 17:51:23 -08:00
  • 4072b5879b
    Merge pull request #2246 from dhiltgen/reject_cuda_without_avx Daniel Hiltgen 2024-01-28 16:26:55 -08:00
  • 15562e887d Don't disable GPUs on arm without AVX Daniel Hiltgen 2024-01-28 15:22:38 -08:00
  • f2245c7c77
    print prompt with OLLAMA_DEBUG=1 (#2245) Jeffrey Morgan 2024-01-28 15:22:35 -08:00
  • e4b9b72f2a
    Do not repeat system prompt for chat templating (#2241) Jeffrey Morgan 2024-01-28 14:15:56 -08:00
  • 311f8e0c3f
    Merge pull request #2243 from dhiltgen/harden_zero_gpus Daniel Hiltgen 2024-01-28 13:30:44 -08:00
  • f07f8b7a9e Harden for zero detected GPUs Daniel Hiltgen 2024-01-28 13:13:10 -08:00
  • 4c4c730a0a
    Merge branch 'ollama:main' into main mraiser 2024-01-27 21:56:11 -05:00
  • e02ecfb6c8
    Merge pull request #2116 from dhiltgen/cc_50_80 Daniel Hiltgen 2024-01-27 10:28:38 -08:00
  • c8059b4dcf
    Merge pull request #2224 from jaglinux/fix_rocm_get_version_message Daniel Hiltgen 2024-01-27 07:29:32 -08:00
  • 59d87127f5
    Update gpu_info_rocm.c Jagadish Krishnamoorthy 2024-01-26 22:08:27 -08:00
  • b5cf31b460
    add keep_alive to generate/chat/embedding api endpoints (#2146) Patrick Devine 2024-01-26 14:28:02 -08:00
  • cc4915e262
    Merge pull request #2214 from dhiltgen/reject_cuda_without_avx Daniel Hiltgen 2024-01-26 12:06:44 -08:00
  • 667a2ba18a Detect lack of AVX and fallback to CPU mode Daniel Hiltgen 2024-01-26 11:11:09 -08:00
  • e054ebe059
    Merge pull request #2212 from ollama/mxyng/fix-build Michael Yang 2024-01-26 11:19:08 -08:00
  • 9d3dcfd0ec fix logging Michael Yang 2024-01-26 11:04:27 -08:00
  • 6e0ea5ecc8
    Merge pull request #1916 from ollama/mxyng/inactivity-monitor Michael Yang 2024-01-26 10:56:00 -08:00
  • a47d8b2557
    Merge pull request #2197 from dhiltgen/remove_rocm_image Daniel Hiltgen 2024-01-26 09:34:23 -08:00
  • 30c43c285c
    Merge pull request #2195 from dhiltgen/rocm_real_gpus Daniel Hiltgen 2024-01-26 09:30:24 -08:00
  • 23a7ea593b
    Merge pull request #2209 from dhiltgen/harden_mgmt Daniel Hiltgen 2024-01-26 09:30:13 -08:00
  • 75c44aa319 Add back ROCm container support Daniel Hiltgen 2024-01-25 16:58:05 -08:00
  • 9d7b5d6c91 Ignore AMD integrated GPUs Daniel Hiltgen 2024-01-25 15:57:32 -08:00
  • 5d9c4a5f5a Fix crash on cuda ml init failure Daniel Hiltgen 2024-01-26 09:18:33 -08:00
  • 197e420a97
    Merge pull request #2196 from dhiltgen/remove_rocm_image Daniel Hiltgen 2024-01-25 16:50:32 -08:00
  • a34e1ad3cf Switch back to ubuntu base Daniel Hiltgen 2024-01-25 16:46:01 -08:00
  • 2ae0556292
    Merge pull request #1679 from ollama/mxyng/build-gpus Michael Yang 2024-01-25 16:38:14 -08:00
  • 5be9bdd444
    Update modelfile.md Jeffrey Morgan 2024-01-25 16:29:48 -08:00
  • b706794905
    Update modelfile.md to include MESSAGE Jeffrey Morgan 2024-01-25 16:29:32 -08:00
  • a8c5413d06 only generate gpu libs Michael Yang 2024-01-19 09:20:19 -08:00
  • 5580de4571 archive ollama binaries Michael Yang 2023-12-22 16:06:35 -08:00
  • 946431d5b0 build cuda and rocm Michael Yang 2023-12-22 12:17:37 -08:00
  • 0610126049 remove env setting Michael Yang 2024-01-18 17:19:12 -08:00
  • 3ebd6a83fc update submodule to cd4fddb29f81d6a1f6d51a0c016bc6b486d68def Jeffrey Morgan 2024-01-25 13:54:11 -08:00
  • a64570dcae
    Fix clearing kv cache between requests with the same prompt (#2186) Jeffrey Morgan 2024-01-25 13:46:20 -08:00
  • 7c40a67841
    Save and load sessions (#2063) Patrick Devine 2024-01-25 12:12:36 -08:00
  • e64b5b07a2
    Merge pull request #2181 from ollama/mxyng/stub-lint Michael Yang 2024-01-25 11:55:15 -08:00
  • 9e1e295cdc
    Merge pull request #2175 from ollama/mxyng/refactor-tensor-read Michael Yang 2024-01-25 09:22:42 -08:00
  • 6eb3cddcb6 To build on NixOS: nix-shell --run 'go generate ./... && go build .' Marc Raiser 2024-01-25 10:17:22 -05:00
  • a4564232a4
    Update gen_linux.sh to find libcudart in separate directory mraiser 2024-01-25 09:49:35 -05:00
  • a643823f86
    Update README.md Jeffrey Morgan 2024-01-24 21:36:56 -08:00
  • 8e5d359a03 stub generate outputs for lint Michael Yang 2024-01-24 17:29:47 -08:00
  • a170888dd4
    Merge pull request #2174 from dhiltgen/rocm_real_gpus Daniel Hiltgen 2024-01-24 11:09:17 -08:00
  • cd22855ef8 refactor tensor read Michael Yang 2024-01-24 10:48:31 -08:00
  • 013fd07139 More logging for gpu management Daniel Hiltgen 2024-01-24 10:32:00 -08:00
  • f63dc2db5c
    Merge pull request #2162 from dhiltgen/rocm_real_gpus Daniel Hiltgen 2024-01-23 17:45:40 -08:00
  • eaa5a396d9
    Update README.md Jeffrey Morgan 2024-01-23 16:08:15 -08:00
  • 8ed22f5d72
    Update README.md Jeffrey Morgan 2024-01-23 14:38:01 -08:00
  • 987c16b2f7 Report more information about GPUs in verbose mode Daniel Hiltgen 2024-01-22 16:03:32 -08:00
  • 950f636d64
    Update README.md Jeffrey Morgan 2024-01-23 10:29:10 -08:00
  • 4458efb73a
    Load all layers on arm64 macOS if model is small enough (#2149) Jeffrey Morgan 2024-01-22 17:40:06 -08:00
  • ceea599494
    Merge pull request #2150 from dhiltgen/default_version Daniel Hiltgen 2024-01-22 17:38:27 -08:00
  • 3005ec74b3 Set a default version using git describe Daniel Hiltgen 2024-01-22 17:12:20 -08:00
  • 0759d8996e
    Merge pull request #2148 from dhiltgen/intel_mac Daniel Hiltgen 2024-01-22 16:56:58 -08:00
  • 0f5b843319 Refine Accelerate usage on mac Daniel Hiltgen 2024-01-22 16:25:56 -08:00
  • ffaf52e1e9 update submodule to 011e8ec577fd135cbc02993d3ea9840c516d6a1c Jeffrey Morgan 2024-01-22 15:16:47 -08:00
  • 940b10b036
    Merge pull request #2144 from jmorganca/mxyng/update-faq Michael Yang 2024-01-22 13:46:57 -08:00
  • 3bc28736cd
    Merge pull request #2143 from dhiltgen/llm_verbosity Daniel Hiltgen 2024-01-22 13:19:16 -08:00
  • 93a756266c faq: update to use launchctl setenv Michael Yang 2024-01-22 12:30:58 -08:00
  • a0a829bf7a
    Merge pull request #2142 from dhiltgen/debug_on_fail Daniel Hiltgen 2024-01-22 12:29:22 -08:00
  • 730dcfcc7a Refine debug logging for llm Daniel Hiltgen 2024-01-22 12:26:49 -08:00
  • 27a2d5af54 Debug logging on init failure Daniel Hiltgen 2024-01-22 12:08:22 -08:00
  • 5f81a33f43
    update submodule to 6f9939d (#2115) Jeffrey Morgan 2024-01-22 11:56:40 -08:00
  • 6225fde046
    Merge pull request #2102 from jmorganca/mxyng/fix-create-override Michael Yang 2024-01-22 09:37:48 -08:00
  • 069184562b
    readline: drop not use min function (#2134) Meng Zhuo 2024-01-23 00:15:08 +08:00
  • 5576bb2348
    Merge pull request #2130 from dhiltgen/more_faster Daniel Hiltgen 2024-01-21 16:14:12 -08:00
  • 2738837786
    Merge pull request #2131 from dhiltgen/probe_cards_at_init Daniel Hiltgen 2024-01-21 16:13:47 -08:00
  • ec3764538d Probe GPUs before backend init Daniel Hiltgen 2024-01-21 15:39:59 -08:00
  • df54c723ae Make CPU builds parallel and customizable AMD GPUs Daniel Hiltgen 2024-01-21 12:57:13 -08:00
  • fa8c990e58
    Merge pull request #2127 from dhiltgen/rocm_container Daniel Hiltgen 2024-01-21 11:49:01 -08:00
  • da72235ebf Combine the 2 Dockerfiles and add ROCm Daniel Hiltgen 2024-01-21 11:37:11 -08:00
  • 89c4aee29e
    Unlock mutex when failing to load model (#2117) Jeffrey Morgan 2024-01-20 20:54:46 -05:00
  • a447a083f2 Add compute capability 5.0, 7.5, and 8.0 Daniel Hiltgen 2024-01-20 12:15:50 -08:00
  • f32ea81b21
    increase minimum overhead to 1024MiB (#2114) Jeffrey Morgan 2024-01-20 17:11:38 -05:00
  • 681a914990 Add support for CUDA 5.2 cards Daniel Hiltgen 2024-01-20 10:48:43 -08:00
  • 4c54f0ddeb
    sign dylibs on macOS (#2101) Jeffrey Morgan 2024-01-19 19:24:11 -05:00
  • c08dfaa23d fix: remove overwritten model layers Michael Yang 2024-01-19 14:58:36 -08:00
  • 3b76e736ae
    Merge pull request #2100 from dhiltgen/more_wsl_globs Daniel Hiltgen 2024-01-19 13:41:08 -08:00
  • 552db98bf1 More WSL paths Daniel Hiltgen 2024-01-19 13:23:29 -08:00
  • fdcdfef620
    Merge pull request #2099 from dhiltgen/fix_cuda_model_swap Daniel Hiltgen 2024-01-19 12:22:04 -08:00
  • 6a042438af Switch to local dlopen symbols Daniel Hiltgen 2024-01-19 11:37:02 -08:00
  • dc88cc3981
    use gzip for runner embedding (#2067) Jeffrey Morgan 2024-01-19 13:23:03 -05:00
  • 62976087c6
    Merge pull request #1999 from lainedfles/termux_android_cpu_only Daniel Hiltgen 2024-01-18 17:16:53 -08:00
  • 344342abdf Restore dyn_ext_server.c since RTLD_DEEPBIND has been removed Self Denial 2024-01-18 17:30:42 -07:00
  • eb76f3e379 Fix CPU-only build under Android Termux enviornment. Self Denial 2024-01-15 02:37:44 -07:00
  • d017e3d0a6
    Merge pull request #2060 from jmorganca/mxyng/fix-show Michael Yang 2024-01-18 16:02:27 -08:00
  • aac9ab4db7 fix show handler Michael Yang 2024-01-18 15:36:50 -08:00
  • 1f5b7ff976
    Merge pull request #1932 from jmorganca/mxyng/api-fields Michael Yang 2024-01-18 14:56:51 -08:00
  • e299831e2c
    Merge pull request #1958 from purificant/ci Michael Yang 2024-01-18 14:53:36 -08:00
  • 745b5934fa add model to ModelResponse Michael Yang 2024-01-18 14:32:55 -08:00
  • a38d88d828 api: add model for all requests Michael Yang 2024-01-11 14:07:54 -08:00
  • abec7f06e5
    Merge pull request #2056 from dhiltgen/slog Daniel Hiltgen 2024-01-18 14:27:24 -08:00
  • e5da190bac
    Merge pull request #2020 from jmorganca/mxyng/install-fedora Michael Yang 2024-01-18 14:23:42 -08:00