Commit Graph

  • 42998d797d
    subprocess llama.cpp server (#401) Bruce MacDonald 2023-08-30 16:35:03 -04:00
  • 924cd9d405 Update development.md Bruce MacDonald 2023-08-30 16:32:58 -04:00
  • d6ca778574 use cmd context cancel Bruce MacDonald 2023-08-30 16:27:24 -04:00
  • 531ca9089e check llama.cpp server running Bruce MacDonald 2023-08-30 16:03:40 -04:00
  • 04f2d03adb set server file by default Bruce MacDonald 2023-08-30 11:29:43 -04:00
  • 197b8ff9c4 windows path changes Bruce MacDonald 2023-08-30 11:18:27 -04:00
  • b84a18f2dc windows cpu build fixes Bruce MacDonald 2023-08-30 11:05:04 -04:00
  • ab836b73eb attempt to support windows cpu builds Bruce MacDonald 2023-08-29 18:13:47 -04:00
  • 5a99c41823 PR feedback Bruce MacDonald 2023-08-29 16:34:07 -04:00
  • 50258f18b1 fix comment typo Bruce MacDonald 2023-08-29 10:59:26 -04:00
  • 8a6d038f08 Update llm/ggml_llama.go Bruce MacDonald 2023-08-28 18:22:23 -04:00
  • a6ed0f5524 generate git patches Bruce MacDonald 2023-08-29 10:47:23 -04:00
  • 3e05677032 apply patches Bruce MacDonald 2023-08-28 18:14:33 -04:00
  • 7df932c78a pr feedback Bruce MacDonald 2023-08-28 18:03:12 -04:00
  • 11489b6968 34b model type Bruce MacDonald 2023-08-28 16:36:33 -04:00
  • ce097b0b21 Update ggml_llama.go Bruce MacDonald 2023-08-28 16:26:06 -04:00
  • 2e563e38b3 Update ggml_llama.go Bruce MacDonald 2023-08-28 14:31:07 -04:00
  • 1128256878 Update ggml_llama.go Bruce MacDonald 2023-08-28 14:28:08 -04:00
  • eb977744e1 update submodule command Bruce MacDonald 2023-08-28 14:26:44 -04:00
  • 56e5cdfeef trim prompt in ollama to return context Bruce MacDonald 2023-08-28 14:22:07 -04:00
  • 5a436adaa9 apply ggml patches Bruce MacDonald 2023-08-28 12:15:22 -04:00
  • a27bcd6139 write on init Bruce MacDonald 2023-08-25 17:15:17 -07:00
  • c29b3b53d4 build gpu and cpu for mac Bruce MacDonald 2023-08-25 17:02:46 -07:00
  • 79ed534a82 random port with retry Bruce MacDonald 2023-08-25 15:27:26 -07:00
  • 5b97daa71a Update llm/llama.go Bruce MacDonald 2023-08-25 09:38:34 -07:00
  • 98a3419c37 Update types.go Bruce MacDonald 2023-08-24 15:40:46 -07:00
  • 1cfd5a4b4a Update llama.go Bruce MacDonald 2023-08-23 14:44:50 -07:00
  • 4bc10683f7 Update development.md Bruce MacDonald 2023-08-23 14:43:24 -07:00
  • ba3aaf3934 remove cpi/gpu fallback Bruce MacDonald 2023-08-23 14:42:45 -07:00
  • aea71cfb1c Update index.ts Bruce MacDonald 2023-08-23 14:12:14 -07:00
  • 0dd0caad9e Update .gitignore Bruce MacDonald 2023-08-23 14:03:26 -07:00
  • bbacd74c19 add submodule to docs Bruce MacDonald 2023-08-23 14:00:57 -07:00
  • d961f5dab8 git submodule llama.cpp Bruce MacDonald 2023-08-23 13:56:13 -07:00
  • aa2586c8c8 remove unused Bruce MacDonald 2023-08-23 13:35:41 -07:00
  • bbb14b6bed Update development.md Bruce MacDonald 2023-08-23 13:29:38 -07:00
  • 334c217ff0 use go generate to get libraries Bruce MacDonald 2023-08-23 13:18:52 -07:00
  • 7267851ef7 remove debug log Bruce MacDonald 2023-08-22 17:30:48 -07:00
  • 553edb882e Update llama.go Bruce MacDonald 2023-08-22 17:12:00 -07:00
  • e72560ea42 remove todo Bruce MacDonald 2023-08-22 17:05:12 -07:00
  • ae6a3a724a clean up Bruce MacDonald 2023-08-22 17:04:24 -07:00
  • 5858346c0b remove sample count and duration metrics Bruce MacDonald 2023-08-22 16:58:24 -07:00
  • 71a68d495d restore prompt num keep Bruce MacDonald 2023-08-22 16:44:21 -07:00
  • a215aabe5a multiple runners Bruce MacDonald 2023-08-22 16:36:31 -07:00
  • d34da4104a stop llama runner when app stops Bruce MacDonald 2023-08-22 16:32:47 -07:00
  • 1bb681ca36 let llama_cpp decide the number of threads to use Bruce MacDonald 2023-08-22 15:33:28 -07:00
  • e37f2e89e2 lora Bruce MacDonald 2023-08-22 14:27:56 -07:00
  • a3aec93fff use request context for llama_cpp Bruce MacDonald 2023-08-22 13:59:51 -07:00
  • d370f5bc42 fix params Bruce MacDonald 2023-08-22 13:03:10 -07:00
  • cbfcf551c6 pack llama.cpp Bruce MacDonald 2023-08-22 12:11:00 -07:00
  • 13364ad5c2 remove c code Bruce MacDonald 2023-08-21 18:03:27 -07:00
  • e2cb384c92 prototype Bruce MacDonald 2023-08-21 17:20:35 -07:00
  • f4432e1dba
    treat stop as stop sequences, not exact tokens (#442) v0.0.17 Quinn Slack 2023-08-30 10:53:42 -05:00
  • 982c535428
    Merge pull request #428 from jmorganca/mxyng/upload-chunks Michael Yang 2023-08-30 07:47:17 -07:00
  • 33ae533c88 treat stop as stop sequences, not exact tokens Quinn Slack 2023-08-29 23:21:53 -05:00
  • e5137e2f6c build: add Docker Compose file and service for running Ollama with Docker blogbin 2023-08-29 21:35:45 +08:00
  • 7df342a6ea
    Merge pull request #421 from jmorganca/mxyng/f16-metal Michael Yang 2023-08-29 06:32:59 -07:00
  • 8bbff2df98
    add model IDs (#439) Patrick Devine 2023-08-28 20:50:24 -07:00
  • a44b7a7fc6 add model IDs Patrick Devine 2023-08-28 20:31:00 -07:00
  • 16b06699fd remove unused parameter Michael Yang 2023-08-28 18:35:18 -04:00
  • 246dc65417 loosen http status code checks Michael Yang 2023-08-26 21:55:21 -07:00
  • 865fceb73c chunked pipe Michael Yang 2023-08-26 08:28:35 -07:00
  • 72266c7684 bump chunk size to 95MB Michael Yang 2023-08-25 15:38:39 -07:00
  • d3b838ce60 update orca to orca-mini Jeffrey Morgan 2023-08-27 13:26:30 -04:00
  • e639a12fa1
    Merge pull request #412 from jmorganca/mxyng/update-readme Michael Yang 2023-08-26 21:26:34 -07:00
  • e82fcf30c6
    Merge pull request #420 from jmorganca/mxyng/34b-mem-check Michael Yang 2023-08-26 14:15:52 -07:00
  • 495e8b0a6a
    Merge pull request #426 from jmorganca/default-template Michael Yang 2023-08-26 14:15:38 -07:00
  • 59734ca24d set default template Michael Yang 2023-08-26 12:20:28 -07:00
  • 22ab7f5f88 default host to 127.0.0.1, fixes #424 Jeffrey Morgan 2023-08-26 11:59:28 -07:00
  • b25dd1795d allow F16 to use metal Michael Yang 2023-08-26 08:33:03 -07:00
  • 304f2b6c96 add 34b to mem check Michael Yang 2023-08-26 08:29:21 -07:00
  • 2ecc3a33c3
    delete all models (not just 1st) in ollama rm (#415) Quinn Slack 2023-08-26 00:47:56 -07:00
  • ff9a93a775 treat ollama run model < file as entire prompt, not prompt-per-line Quinn Slack 2023-08-25 21:54:52 -07:00
  • f728d44737 delete all models (not just 1st) in ollama rm Quinn Slack 2023-08-25 21:35:02 -07:00
  • ee6e1df118 add codellama to model list in readme Jeffrey Morgan 2023-08-25 20:44:26 -07:00
  • 177b69a211 add missing entries for 34B v0.0.16 Jeffrey Morgan 2023-08-25 18:35:35 -07:00
  • dad63f0821
    Merge pull request #411 from jmorganca/mxyng/34b Michael Yang 2023-08-25 11:59:05 -07:00
  • 041f9ad1a1 update README.md Michael Yang 2023-08-25 11:44:25 -07:00
  • 7a378f8b66 patch llama.cpp for 34B Michael Yang 2023-08-25 10:06:55 -07:00
  • de0bdd7f29
    Merge pull request #405 from jmorganca/mxyng/34b Michael Yang 2023-08-24 10:37:22 -07:00
  • b1cececb8e add 34b model type Michael Yang 2023-08-24 10:35:44 -07:00
  • 8a36ec6e4a decode gguf Michael Yang 2023-08-22 15:49:39 -07:00
  • e0d39fa3bf
    Merge pull request #398 from jmorganca/mxyng/cleanup Michael Yang 2023-08-22 15:51:41 -07:00
  • 968ced2e71
    Merge pull request #393 from jmorganca/mxyng/net-url Michael Yang 2023-08-22 15:51:33 -07:00
  • 32d1a00017 remove unused requestContextKey Michael Yang 2023-08-22 08:51:21 -07:00
  • 04e2128273 move upload funcs to upload.go Michael Yang 2023-08-22 08:50:21 -07:00
  • 2cc634689b use url.URL Michael Yang 2023-08-21 18:38:31 -07:00
  • 8f827641b0
    Merge pull request #397 from jmorganca/mxyng/release-mode Michael Yang 2023-08-22 10:48:44 -07:00
  • 95187d7e1e build release mode Michael Yang 2023-08-22 09:48:35 -07:00
  • 9ec7e37534
    Merge pull request #392 from jmorganca/mxyng/version Michael Yang 2023-08-22 09:50:25 -07:00
  • 2c7f956b38 add version Michael Yang 2023-08-21 18:24:42 -07:00
  • a9f6c56652 fix FROM instruction erroring when referring to a file Jeffrey Morgan 2023-08-22 09:39:42 -07:00
  • 4d9aa37e46 Document what happens upon first app launch Justin Mayer 2023-08-22 10:59:16 +02:00
  • 0a892419ad
    Strip protocol from model path (#377) Ryan Baker 2023-08-21 21:56:56 -07:00
  • 1178fd2cbb build with cmake shell Jeffrey Morgan 2023-08-21 18:36:21 -07:00
  • e3054fc74e add .env to .dockerignore Jeffrey Morgan 2023-08-21 09:32:02 -07:00
  • 97c15b601a wip shell Jeffrey Morgan 2023-08-21 09:24:22 -07:00
  • 23c2485044
    Merge pull request #381 from jmorganca/mxyng/fix-push-chunks Michael Yang 2023-08-18 13:49:25 -07:00
  • 386c66f285
    Merge pull request #378 from jmorganca/mxyng/copy-metadata-from-source Michael Yang 2023-08-18 13:49:09 -07:00
  • 06f8f01e38
    Closes #371 jesjess243 2023-08-18 14:32:55 -04:00
  • cc06153757
    Closes #371 jesjess243 2023-08-18 14:29:33 -04:00