Commit Graph

  • 44f7c4f568 give direction to user when runner fails Bruce MacDonald 2023-10-06 12:50:58 -04:00
  • 77295f716e
    prevent waiting on exited command (#752) Bruce MacDonald 2023-10-11 12:32:13 -04:00
  • c7d4f90f17 Update llama.go Bruce MacDonald 2023-10-11 12:31:30 -04:00
  • cdfe62122f close llama runner once Bruce MacDonald 2023-10-11 11:22:29 -04:00
  • c188699884 feat: add network discovery via zeroconf Eric allen 2023-10-10 17:20:52 -04:00
  • 615f7d1dea cleanup readme. Matt Williams 2023-10-11 06:13:29 -07:00
  • cdf5e106ae rename dirs Matt Williams 2023-10-11 06:10:24 -07:00
  • a85329f59a rename the models to be more descriptive Matt Williams 2023-10-10 17:40:02 -07:00
  • 1c0f7cbdae prevent waiting on exited command Bruce MacDonald 2023-10-10 18:02:23 -04:00
  • f2ba1311aa
    improve vram safety with 5% vram memory buffer (#724) Bruce MacDonald 2023-10-10 16:16:09 -04:00
  • e11454c050 wait for command to exit, no timeout Bruce MacDonald 2023-10-10 13:37:32 -04:00
  • 65dcd0ce35
    always cleanup blob download (#747) Jeffrey Morgan 2023-10-10 13:12:29 -04:00
  • 0040f543a2
    Merge pull request #743 from jmorganca/mxyng/http-proxy Michael Yang 2023-10-10 09:59:06 -07:00
  • 5048f1beb1 logging format Bruce MacDonald 2023-10-10 12:57:51 -04:00
  • be707cb893 wait for subprocess to exit Bruce MacDonald 2023-10-10 12:53:55 -04:00
  • 01c04a4621 rename variable Bruce MacDonald 2023-10-10 11:33:52 -04:00
  • 36c4681f09 always cleanup blob download Jeffrey Morgan 2023-10-09 23:10:33 -04:00
  • 130a617607 new ollama Daniyel Yaacov 2023-10-09 17:41:33 -04:00
  • 767f9bdbbb
    Merge pull request #585 from jmorganca/matt/examplementors Matt Williams 2023-10-09 13:58:14 -07:00
  • f7f5169c94
    Update api.md (#741) Costa Alexoglou 2023-10-09 22:01:46 +02:00
  • 2cfffea02e handle client proxy Michael Yang 2023-10-09 12:18:26 -07:00
  • f6e98334e4 handle upstream proxies Michael Yang 2023-10-09 11:42:36 -07:00
  • 8dd0e4b853
    Update api.md Costa Alexoglou 2023-10-09 17:08:28 +02:00
  • d260a528ab
    Update README.md Timothy Jaeryang Baek 2023-10-09 00:52:49 -05:00
  • 569026779b layout updated Timothy J. Baek 2023-10-08 19:58:43 -07:00
  • 83ae37f154 chat event stream behaviour updated Timothy J. Baek 2023-10-08 19:42:54 -07:00
  • 568ce38fa1 endpoint address updated Timothy J. Baek 2023-10-08 18:32:54 -07:00
  • c4413a7afa edited typo Timothy J. Baek 2023-10-08 16:29:05 -07:00
  • 9e30abc16a ollama placeholder added Timothy J. Baek 2023-10-08 15:58:33 -07:00
  • 9a220ede61
    Update LICENSE Timothy Jaeryang Baek 2023-10-08 17:42:48 -05:00
  • 5e03670f1e chat feature added Timothy J. Baek 2023-10-08 15:38:42 -07:00
  • 5cd4946df2
    Initial commit Timothy Jaeryang Baek 2023-10-06 17:08:27 -05:00
  • ab0668293c llm: fix build on amd64 Jeffrey Morgan 2023-10-06 14:39:54 -07:00
  • 805c7d8d47 5% buffer rather than 10% Bruce MacDonald 2023-10-06 17:30:13 -04:00
  • 9b8afa1afb check free memory not total Bruce MacDonald 2023-10-06 17:24:57 -04:00
  • 74f5c4aa84 improve vram safety Bruce MacDonald 2023-10-06 16:21:30 -04:00
  • af4cf55884
    not found error before pulling model (#718) Bruce MacDonald 2023-10-06 16:06:20 -04:00
  • d6786f2945
    add feedback for reading model metadata (#722) Bruce MacDonald 2023-10-06 16:05:32 -04:00
  • 38dc2f79bc
    Merge pull request #626 from jmorganca/mxyng/concurrent-downloads Michael Yang 2023-10-06 13:01:29 -07:00
  • cb961c87ca
    Merge pull request #679 from jamesbraza/modelfile-docs Michael Yang 2023-10-06 12:59:45 -07:00
  • 0560b28a8d names Michael Yang 2023-10-03 17:06:13 -07:00
  • 10199c5987 replace done channel with file check Michael Yang 2023-10-03 16:52:49 -07:00
  • 288814d3e4 fix ref counts Michael Yang 2023-10-03 16:44:35 -07:00
  • 04733438da check head request response Michael Yang 2023-10-03 16:12:53 -07:00
  • 711e891f0f fix resumable downloads Michael Yang 2023-10-02 15:26:27 -07:00
  • 090d08422b handle unexpected eofs Michael Yang 2023-10-02 13:34:07 -07:00
  • 5b84404c64 handle concurrent requests for the same blobs Michael Yang 2023-09-29 16:13:53 -07:00
  • 8544edca21 parallel chunked downloads Michael Yang 2023-09-27 16:22:30 -07:00
  • 77a9a1174b add feedback for reading model metadata Bruce MacDonald 2023-10-06 14:16:46 -04:00
  • 0d9da05bcd not found error before pulling model Bruce MacDonald 2023-10-06 11:11:23 -04:00
  • 5d22319a2c
    rename server subprocess (#700) Bruce MacDonald 2023-10-06 10:15:42 -04:00
  • e613cd8ad0
    Update url for Swagger editor Mehdi 2023-10-05 15:21:19 -07:00
  • 7694ca8486
    add API readme.md Mehdi 2023-10-05 14:25:06 -07:00
  • 3124dcecaa
    Create readme.md Mehdi 2023-10-05 14:20:15 -07:00
  • e069b2b9f7 OAS3.0 API Spec marscod 2023-10-05 14:13:44 -07:00
  • 2130c0708b
    output type parsed from modelfile (#678) Bruce MacDonald 2023-10-05 14:58:04 -04:00
  • 413a9155e2 validate api options fields from map Bruce MacDonald 2023-10-05 14:52:10 -04:00
  • 61ff1946e6
    revise help text (#706) Patrick Devine 2023-10-05 11:36:07 -07:00
  • d06bc0cb6e
    enable q8, q5, 5_1, and f32 for linux gpu (#699) Bruce MacDonald 2023-10-05 12:53:47 -04:00
  • 76a965fe12 display message if the model take a while to load Bruce MacDonald 2023-10-05 12:50:05 -04:00
  • 37ae8cde83 Revert "async model preload" Bruce MacDonald 2023-10-05 12:32:47 -04:00
  • 4e5bb5472c Update gguf to latest Bruce MacDonald 2023-10-05 11:16:06 -04:00
  • d104b7e997
    Fix go test./... issue: fmt.Println arg list ends with redundant newline (#705) Alexander F. Rødseth 2023-10-05 17:11:04 +02:00
  • d206ef7fd8 Use Go 1.21 in the Dockerfile Alexander F. Rødseth 2023-10-05 09:41:46 +02:00
  • 9c7d8376b5 revise help text Patrick Devine 2023-10-04 17:29:28 -07:00
  • f17499967e Fix go test./... issue: fmt.Println arg list ends with redundant newline Alexander F. Rødseth 2023-10-04 23:58:57 +02:00
  • 500879857d async model preload Bruce MacDonald 2023-10-04 16:44:43 -04:00
  • 549a26efd9 windows fix Bruce MacDonald 2023-10-04 15:49:31 -04:00
  • d156ee6292 rename server subprocess Bruce MacDonald 2023-10-04 15:32:49 -04:00
  • 9e2de1bd2c
    increase streaming buffer size (#692) Bruce MacDonald 2023-10-04 14:09:00 -04:00
  • 9907460e40 enable q8, q5, 5_1, and f32 for linux gpu Bruce MacDonald 2023-10-03 18:26:06 -04:00
  • 903d4f1dee increase streaming buffer size Bruce MacDonald 2023-10-03 16:24:32 -04:00
  • dc87e9c9ae update Dockerfile to pass GOFLAGS Jeffrey Morgan 2023-10-03 07:05:09 -07:00
  • 367cb68dc1
    Merge pull request #686 from jmorganca/mxyng/starcoder Michael Yang 2023-10-02 22:47:19 -07:00
  • c02c0cd483 starcoder Michael Yang 2023-10-02 19:52:25 -07:00
  • 1852755154
    show a default message when license/parameters/system prompt/template aren't specified (#681) v0.1.1 Patrick Devine 2023-10-02 14:34:52 -07:00
  • 47d930badf show a default message when license/parameters/system prompt/template aren't specified Patrick Devine 2023-10-02 14:26:08 -07:00
  • 6f2ce74231 Got rif of all caps to show it can be lower case James Braza 2023-10-02 13:54:27 -07:00
  • 6edcc5c79f Using code highlighting syntax around Modelfile James Braza 2023-10-02 13:46:05 -07:00
  • 23791d7b03 output type parsed from modelfile Bruce MacDonald 2023-10-02 16:42:03 -04:00
  • b1f7123301
    clean up num_gpu calculation code (#673) Bruce MacDonald 2023-10-02 14:53:42 -04:00
  • 1fbf3585d6
    Relay default values to llama runner (#672) Bruce MacDonald 2023-10-02 14:53:16 -04:00
  • a113624db0 omit empty stop Bruce MacDonald 2023-10-02 14:52:12 -04:00
  • 99d5161e8a
    don't wordwrap when stdout is redirected or piped (#662) Patrick Devine 2023-10-02 11:50:55 -07:00
  • 6ba02bbe03 Update llama.go Bruce MacDonald 2023-10-02 14:12:29 -04:00
  • 4d57ae595e clean up num_gpu calculation code Bruce MacDonald 2023-10-02 14:11:09 -04:00
  • 900e5ddb1a relay default predict options to llama.cpp Bruce MacDonald 2023-10-02 13:29:52 -04:00
  • f29a7db98e include seed in params for llama.cpp server and remove empty filter for temp hallh 2023-10-01 13:09:57 +02:00
  • ea8380be45
    add community project: Chatbot Ollama Michael 2023-10-02 09:04:31 -07:00
  • 75d3de4b66
    Merge 4b15b5c46bb0c54763994004e5461ab289315bce into 4f25092dc13d84bb10463e1e7ad818b20a4e1846 Quinn Slack 2023-10-01 22:18:17 -07:00
  • 4f25092dc1 fix build_docker.sh permissions Jeffrey Morgan 2023-10-01 16:42:32 -07:00
  • 949fc4eafa wip /api/chat api Jeffrey Morgan 2023-10-01 14:54:17 -07:00
  • 4fc10acce9
    add some missing code directives in docs (#664) Jiayu Liu 2023-10-02 02:51:01 +08:00
  • 343ca755dd
    add some missing code directives in docs Jiayu Liu 2023-10-01 21:24:12 +08:00
  • 6a37640582 include seed in params for llama.cpp server and remove empty filter for temp hallh 2023-10-01 13:09:57 +02:00
  • 77e1adf80e don't wordwrap when stdout is redirected or piped Patrick Devine 2023-09-30 16:26:17 -07:00
  • 730872c434
    Added Docker hub link James Braza 2023-09-30 15:08:00 -07:00
  • 050215b0c1
    Documenting OpenAI compatibility James Braza 2023-09-30 15:02:28 -07:00
  • 1c1b0f67bf
    Combined brew install lines James Braza 2023-09-30 14:59:14 -07:00
  • 0a4f21c0a7
    fix docker build (#659) Michael Yang 2023-09-30 13:34:01 -07:00