Commit Graph

  • 5d71bda478
    close llm on interrupt (#577) Bruce MacDonald 2023-09-22 19:41:52 +01:00
  • 88897a90e4 fix ipv6 parse ip Michael Yang 2023-09-22 10:40:54 -07:00
  • 9df31c3518
    linux installer script (#534) Bruce MacDonald 2023-09-22 17:01:03 +01:00
  • 2044f9d4da
    Merge pull request #570 from jmorganca/mxyng/head-request Michael Yang 2023-09-21 16:56:17 -07:00
  • 0d186f3b33
    Merge pull request #569 from jmorganca/mxyng/update-submodules Michael Yang 2023-09-21 16:52:42 -07:00
  • 82f5b66c01 register HEAD /api/tags Michael Yang 2023-09-21 16:38:03 -07:00
  • c986694367 fix HEAD / request Michael Yang 2023-09-21 16:35:38 -07:00
  • 058d0cd04b silence warm up log Michael Yang 2023-09-21 14:48:00 -07:00
  • ee1c994d15
    update submodule (#567) Michael Yang 2023-09-21 13:22:23 -07:00
  • 4cba75efc5
    remove tmp directories created by previous servers (#559) Bruce MacDonald 2023-09-21 20:38:49 +01:00
  • 8c83701e9f
    Merge pull request #566 from jmorganca/mxyng/api-check-model-exists Michael Yang 2023-09-21 10:35:14 -07:00
  • 6137b12799 validate existence and pull model using api Michael Yang 2023-09-21 09:50:52 -07:00
  • 1fabba474b refactor default allow origins Michael Yang 2023-09-21 09:42:16 -07:00
  • 765770efdb
    Merge pull request #562 from jmorganca/mxyng/fix-ollama-host Michael Yang 2023-09-20 19:54:47 -07:00
  • 9297ff8330 fix OLLAMA_HOST parsing for ip6 Michael Yang 2023-09-20 17:49:48 -07:00
  • ee4fd16f2c
    Merge pull request #556 from jmorganca/pack-cuda Michael Yang 2023-09-20 15:02:36 -07:00
  • a9ed7cc6aa rename generate.go Michael Yang 2023-09-20 14:39:15 -07:00
  • 6c6a31a1e8 embed libraries using cmake Michael Yang 2023-09-20 12:15:23 -07:00
  • fc6ec356fc remove libcuda.so Bruce MacDonald 2023-09-20 20:36:14 +01:00
  • 1255bc9b45 only package 11.8 runner Bruce MacDonald 2023-09-20 20:00:41 +01:00
  • 084e4c782a
    Merge pull request #557 from jmorganca/mxyng/cleanup Michael Yang 2023-09-20 11:51:01 -07:00
  • 58ffa03d8b fix impossible condition Michael Yang 2023-09-19 15:23:33 -07:00
  • 637f8bc6a5
    Merge pull request #536 from jmorganca/mxyng/redirect-uploads Michael Yang 2023-09-20 11:27:03 -07:00
  • 499e9007a5 pick chunksize based on location Michael Yang 2023-09-19 14:22:54 -07:00
  • b9bb5ca288 use cuda_version Bruce MacDonald 2023-09-20 17:58:16 +01:00
  • 4e8be787c7 pack in cuda libs Bruce MacDonald 2023-09-20 17:40:42 +01:00
  • aa45d7c1df draft: explicitly follow upload redirects Michael Yang 2023-09-14 15:42:50 -07:00
  • e35565c567
    Merge pull request #555 from jmorganca/mxyng/fix-windows-startup Michael Yang 2023-09-19 10:51:58 -07:00
  • a5520bfb42 fix build Michael Yang 2023-09-19 10:42:20 -07:00
  • 2627c464ba
    Merge pull request #554 from jmorganca/mxyng/fix-windows-startup Michael Yang 2023-09-19 09:42:12 -07:00
  • b58d5d16b0 fix mkdir on windows Michael Yang 2023-09-19 09:36:30 -07:00
  • 24580df958
    only add a layer if there is actual data (#535) Patrick Devine 2023-09-18 13:47:45 -07:00
  • 80dd44e80a
    Cmd changes (#541) Patrick Devine 2023-09-18 12:26:56 -07:00
  • 94e1d96b29
    Updated README section on community projects for table (#550) James Braza 2023-09-18 12:22:50 -07:00
  • 66003e1d05
    subprocess improvements (#524) Bruce MacDonald 2023-09-18 15:16:32 -04:00
  • c345053a8b
    Merge pull request #537 from jmorganca/mxyng/upload Michael Yang 2023-09-15 17:48:39 -07:00
  • 08d7c2a944 fix error on upload chunk Michael Yang 2023-09-15 15:59:30 -07:00
  • bc9573dcb1
    Merge pull request #530 from jmorganca/mxyng/progresswriter Michael Yang 2023-09-15 12:43:46 -07:00
  • e53bc57d4d split uploadBlobChunked Michael Yang 2023-09-14 13:30:28 -07:00
  • f0b398d17f implement ProgressWriter Michael Yang 2023-09-14 09:54:05 -07:00
  • 8efbc5df55
    DRAFT: add a simple python client to access ollama (#522) Patrick Devine 2023-09-14 16:37:38 -07:00
  • ccc3e9ac6d
    Merge pull request #531 from jmorganca/mxyng/content-length Michael Yang 2023-09-14 13:33:11 -07:00
  • daa4f096f9 set request.ContentLength Michael Yang 2023-09-14 10:05:29 -07:00
  • 3ee85f1c6c
    Merge pull request #526 from jmorganca/mxyng/cleanup Michael Yang 2023-09-14 13:10:59 -07:00
  • 2540c9181c
    support for packaging in multiple cuda runners (#509) Bruce MacDonald 2023-09-14 15:08:13 -04:00
  • 83ffb154bc
    Merge pull request #507 from jmorganca/mxyng/build Michael Yang 2023-09-14 11:25:59 -07:00
  • 9aa192c812 update cuda docker image Michael Yang 2023-09-08 13:41:49 -07:00
  • fc8707686f
    Update API docs (#527) Matt Williams 2023-09-14 08:51:26 -07:00
  • e2389b63aa add examples of streaming in python and node matt/streamingapi Matt Williams 2023-09-14 07:12:09 -07:00
  • f89c23764b
    Merge pull request #525 from jmorganca/mxyng/falcon-decode Michael Yang 2023-09-13 15:08:47 -07:00
  • e6881cabd0 remove unused Michael Yang 2023-09-13 11:46:29 -07:00
  • d028853879 fix: add falcon.go Michael Yang 2023-09-13 14:47:32 -07:00
  • 949553db23
    Merge pull request #519 from jmorganca/mxyng/decode Michael Yang 2023-09-13 12:43:57 -07:00
  • 0c5a454361 fix model type for 70b Michael Yang 2023-09-12 10:52:57 -07:00
  • f59c4d03f7
    fix ggml arm64 cuda build (#520) Bruce MacDonald 2023-09-12 17:06:48 -04:00
  • 7dee25a07f fix falcon decode Michael Yang 2023-09-12 10:01:20 -07:00
  • f221637053
    first pass at linux gpu support (#454) Bruce MacDonald 2023-09-12 11:04:35 -04:00
  • 45ac07cd02
    create the blobs directory correctly (#508) v0.0.19 Patrick Devine 2023-09-11 14:54:52 -07:00
  • 7d749cc787 fix darwin build script Jeffrey Morgan 2023-09-11 16:31:46 -04:00
  • e7e91cd71c
    add autoprune to remove unused layers (#491) Patrick Devine 2023-09-11 11:46:35 -07:00
  • 3920e15386
    add model format to config layer (#497) Jeffrey Morgan 2023-09-09 17:53:44 -04:00
  • 41e976edde
    Merge pull request #492 from jmorganca/mxyng/nil-pointer Michael Yang 2023-09-07 17:25:23 -07:00
  • de227b620f fix nil pointer dereference Michael Yang 2023-09-07 17:24:31 -07:00
  • 63def6ca49
    Merge pull request #487 from jmorganca/mxyng/dockerignore Michael Yang 2023-09-07 14:16:17 -07:00
  • 738fe9c4aa
    Merge pull request #486 from jmorganca/mxyng/fix-push Michael Yang 2023-09-07 13:58:34 -07:00
  • a8da0bacbe update dockerignore Michael Yang 2023-09-07 13:36:25 -07:00
  • bf146fb072 fix retry on unauthorized chunk Michael Yang 2023-09-07 12:01:50 -07:00
  • f0f4943577 fix get auth token Michael Yang 2023-09-07 11:49:36 -07:00
  • 09dd2aeff9
    GGUF support (#441) Bruce MacDonald 2023-09-07 13:55:37 -04:00
  • 23481167a7 add model_format config var format-config Jeffrey Morgan 2023-09-07 13:54:53 -04:00
  • 07b4074e7b
    [docs] Improve build instructions (#482) Alexander Pepper 2023-09-07 12:43:26 +02:00
  • 61dda6a5e0 set minimum CMAKE_OSX_DEPLOYMENT_TARGET to 11.0 Jeffrey Morgan 2023-09-06 19:56:50 -04:00
  • e1f9ced568
    Merge pull request #479 from jmorganca/mxyng/dockerfile Michael Yang 2023-09-06 15:44:24 -07:00
  • 9795b43d93 update dockerfile Michael Yang 2023-09-06 14:14:20 -07:00
  • 0980d5c7e3
    Merge pull request #478 from jmorganca/mxyng/cleanup Michael Yang 2023-09-06 15:18:54 -07:00
  • 0dae34b6a7 remove unused openssh key types Michael Yang 2023-09-06 14:33:55 -07:00
  • 83c6be1666
    fix model manifests (#477) v0.0.18 Michael Yang 2023-09-06 14:30:08 -07:00
  • 1adfa67589
    tighten up the error string for ollama show flags (#476) Patrick Devine 2023-09-06 13:38:49 -07:00
  • 790d24eb7b
    add show command (#474) Patrick Devine 2023-09-06 11:04:17 -07:00
  • 7de300856b use osPath in gpu check Jeffrey Morgan 2023-09-05 21:52:21 -04:00
  • 213ffdb548 macos amd64 compatibility fixes Jeffrey Morgan 2023-09-05 21:33:27 -04:00
  • d42d88386a
    Merge pull request #473 from jmorganca/mxyng/fix-manifest-path Michael Yang 2023-09-05 17:37:41 -07:00
  • 154f24af91
    Added missing options params to the embeddings docs (#472) Ackermann Yuriy 2023-09-06 12:18:49 +12:00
  • a1ecdd36d5 create manifests directory Michael Yang 2023-09-05 17:10:40 -07:00
  • d18282bfda
    metal: add missing barriers for mul-mat (#469) Bruce MacDonald 2023-09-05 19:37:13 -04:00
  • 9ae76ba8c9
    Merge pull request #471 from jmorganca/mxyng/fix-empty-response Michael Yang 2023-09-05 15:23:05 -07:00
  • 2bc06565c7 fix empty response Michael Yang 2023-09-05 15:03:24 -07:00
  • d1c2558f7e
    Merge pull request #461 from jmorganca/mxyng/fix-inherit-params Michael Yang 2023-09-05 12:30:23 -07:00
  • 7b5aefb427
    Merge pull request #462 from jmorganca/mxyng/rm-marshal-prompt Michael Yang 2023-09-05 11:48:41 -07:00
  • 06ef90c051 fix parameter inheritence Michael Yang 2023-09-05 11:05:03 -07:00
  • 7efbc84320
    Merge pull request #464 from jmorganca/mxyng/fix-num-keep Michael Yang 2023-09-05 11:30:45 -07:00
  • e9f6df7dca use slices.DeleteFunc Michael Yang 2023-09-02 14:38:51 -04:00
  • 7fa6e51686
    generate binary dependencies based on GOARCH on macos (#459) Jeffrey Morgan 2023-09-05 12:53:57 -04:00
  • 8dc68417e7
    Merge pull request #463 from jmorganca/mxyng/fix-last-token Michael Yang 2023-09-05 09:01:32 -07:00
  • cdc9aa14ed template extra args mxyng/extra-args Michael Yang 2023-09-03 18:29:54 -04:00
  • 681f3c4c42 fix num_keep Michael Yang 2023-09-03 17:36:14 -04:00
  • 59a705525c fix not forwarding last token Michael Yang 2023-09-03 17:46:35 -04:00
  • 5d3f314b0b remove marshalPrompt which is no longer needed Michael Yang 2023-09-03 14:10:03 -04:00
  • adaa13088b
    Merge pull request #457 from sqs/dont-html-escape-prompt Michael Yang 2023-09-01 17:41:53 -07:00
  • 62d29b2157 do not HTML-escape prompt Quinn Slack 2023-09-01 17:12:20 -05:00