Commit Graph

  • 86c60f6721 fix docker build Michael Yang 2023-09-30 12:44:09 -07:00
  • 9abb66254a docker: fix volume permission errors Jeffrey Morgan 2023-09-30 12:32:06 -07:00
  • 913eaee26f
    Add colab badge Ibrahim H 2023-09-30 12:56:30 +01:00
  • 1d0ebe67e8
    Document response stream chunk delimiter. (#632) Jay Nakrani 2023-09-30 00:45:52 -04:00
  • 858bfa5611
    Update docs/api.md Jay Nakrani 2023-09-30 00:12:32 -04:00
  • 35695f8691
    Update docs/api.md Jay Nakrani 2023-09-30 00:00:20 -04:00
  • 29d5f2bf2c
    Update docs/api.md Jay Nakrani 2023-09-30 00:00:14 -04:00
  • 081f691ba7
    Tabs 65a 2023-09-29 19:48:26 -07:00
  • 213ebccfc9
    Re-enable CLBlast where appropriate. This should build an accelerated binary for almost everyone. 65a 2023-09-29 19:46:50 -07:00
  • e47f2369e4
    Correct arguments to CMake clean. It shouldn't necessarily be useful in a clean or CI environment, but should make testing locally cleaner. 65a 2023-09-29 19:25:34 -07:00
  • 7b7aa95afe
    Simplify cuda build for linux 65a 2023-09-29 19:19:04 -07:00
  • eac4d43e54
    Create a build script for the cuda binaries. For now just pick between CUBLAS and HIPBLAS 65a 2023-09-29 19:18:02 -07:00
  • a5aa21cd22
    Delete llm/llama.cpp/generate_cuda_linux.go 65a 2023-09-29 19:15:22 -07:00
  • 1e78ce16dc
    Emit a message if a VRAM check failed, but make it clear it's not necessarily bad, we're just going to skip that VRAM check. 65a 2023-09-29 17:48:46 -07:00
  • 94c71b2df1
    Use nvcc presence instead of nvidia-smi 65a 2023-09-29 17:45:06 -07:00
  • a1b2d95f96
    remove unused push/pull params (#650) Bruce MacDonald 2023-09-29 17:27:19 -04:00
  • 2f4fc9c0eb remove unused push/pull params Bruce MacDonald 2023-09-29 17:26:03 -04:00
  • c0b1bf7537
    Merge pull request #606 from jmorganca/mxyng/install.sh-2 Michael Yang 2023-09-29 11:30:46 -07:00
  • cdfeb165ca
    Merge pull request #608 from jmorganca/mxyng/build Michael Yang 2023-09-29 11:30:25 -07:00
  • 92d454ec5f update build_darwin.sh Michael Yang 2023-09-26 10:38:32 -07:00
  • 9333b0cc82
    Merge pull request #612 from jmorganca/mxyng/prune-empty-directories Michael Yang 2023-09-29 11:23:39 -07:00
  • 9771b1ec51
    windows runner fixes (#637) Bruce MacDonald 2023-09-29 11:47:55 -04:00
  • b32cdc91c6 PR feedback Bruce MacDonald 2023-09-29 11:45:59 -04:00
  • 4f8384ac2b embedded path fixes Bruce MacDonald 2023-09-29 11:15:14 -04:00
  • eae2e48704 correct file paths before returning runners Bruce MacDonald 2023-09-29 10:46:46 -04:00
  • 93d7d7e21a Revert "windows runner fixes" Bruce MacDonald 2023-09-29 10:37:24 -04:00
  • 6e3a72001c
    Use llama.cpp generator to enable dynamically picking up SDKs from build environment 65a 2023-09-28 17:27:20 -07:00
  • 7286a87353
    Create runnable but not built go file to generate llama.cpp with options set dynamically based on available SDKs. 65a 2023-09-28 17:26:39 -07:00
  • 4337051350
    Make VRAM reporting support CUDA or ROCm, in that order 65a 2023-09-28 17:24:53 -07:00
  • 76db4a49cf
    allow the user to cancel generating with ctrl-C (#641) Patrick Devine 2023-09-28 17:13:01 -07:00
  • 4ede437bda allow the user to cancel generating with ctrl-C Patrick Devine 2023-09-28 14:47:43 -07:00
  • 3b76c84173 remove list from interactive mode Bruce MacDonald 2023-09-28 17:47:11 -04:00
  • 4aa0976a2e
    Added missing return preventing SIGSEGV because of missing resp (#621) Luc Stepniewski 2023-09-28 23:25:22 +02:00
  • 92c20fdae6
    fix error messages for unknown commands in the repl (#611) Patrick Devine 2023-09-28 14:19:45 -07:00
  • c951da7096
    Merge pull request #634 from jmorganca/mxyng/int64 Michael Yang 2023-09-28 14:17:47 -07:00
  • 2b15d10aaa add cuda to windows generation Bruce MacDonald 2023-09-28 16:14:34 -04:00
  • 6a3552682f windows runner fixes Bruce MacDonald 2023-09-28 16:12:41 -04:00
  • 24d82a23a2
    do not download updates multiple times (#633) Bruce MacDonald 2023-09-28 15:29:17 -04:00
  • 984fb9f4e8 Update index.ts Bruce MacDonald 2023-09-28 15:27:40 -04:00
  • 17152f59ca download refactor Bruce MacDonald 2023-09-28 14:22:50 -04:00
  • f40b3de758 use int64 consistently Michael Yang 2023-09-28 10:00:34 -07:00
  • 5f4008c296
    Update README.md Michael 2023-09-28 09:06:03 -07:00
  • 6ae33d8141
    Update modelfile.md to reflect the usage of num_gpu. (#629) Aaron Coffey 2023-09-28 07:21:21 -07:00
  • c32ec2ff9b do not download updates multiple times Bruce MacDonald 2023-09-28 10:16:58 -04:00
  • a256fb66b1 Document response stream chunk delimiter. Jay Nakrani 2023-09-28 09:23:33 -04:00
  • ca96c61ce9
    Adding mistral 7B to README.md Bhagya Nirmaan Silva 2023-09-28 12:38:54 +02:00
  • 43ab172638
    Update modelfile.md to reflect the usage of num_gpu. Aaron Coffey 2023-09-27 21:06:36 -07:00
  • f6ce879253 Merge remote-tracking branch 'origin/main' Abdullah Ali 2023-09-28 00:46:41 +03:00
  • c5664c1fef
    Update faq.md Jeffrey Morgan 2023-09-27 13:49:43 -07:00
  • 958a5a8184 revert fedora cuda version check Bruce MacDonald 2023-09-27 15:12:29 -04:00
  • 8608eb4760 prune empty directories Michael Yang 2023-09-26 17:28:14 -07:00
  • a2b210130f
    fedora install fixes (#609) Bruce MacDonald 2023-09-27 11:43:47 -04:00
  • 052516d51f Update install.sh Bruce MacDonald 2023-09-27 11:30:13 -04:00
  • ed20837f9a Update modelfile.md Bruce MacDonald 2023-09-27 10:38:10 -04:00
  • 1db2a61dd0
    Added num_predict to the options table (#614) James Braza 2023-09-27 07:26:08 -07:00
  • 2f2a5c20c2 Added missing return preventing SIGSEGV because of missing resp Luc Stepniewski 2023-09-27 12:44:31 +02:00
  • 2ded8ab206 use 11.8.0 nvidia dockerfile base image for now Jeffrey Morgan 2023-09-25 23:52:05 -07:00
  • e6b3648bbf
    Merge pull request #616 from jmorganca/mxyng/fix-model-name Michael Yang 2023-09-26 20:54:18 -07:00
  • 0625e805f0 fix model name not matching Michael Yang 2023-09-26 19:49:55 -07:00
  • 4b15b5c46b add ollama run flags: template, context, stop Quinn Slack 2023-09-26 22:30:52 -04:00
  • e99ec8278b Added num_predict to the options table James Braza 2023-09-26 18:45:32 -07:00
  • 7b5e306149 fix error messages for unknown commands in the repl Patrick Devine 2023-09-26 17:32:21 -07:00
  • c38ec5befb
    Merge pull request #598 from jmorganca/mxyng/help-exit Michael Yang 2023-09-26 15:17:40 -07:00
  • 74aa63fd85 fedora install fixes Bruce MacDonald 2023-09-26 16:36:27 -04:00
  • 6a14fb9277 Allow setting ollama home directory through environment var OLLAMA_HOME. Jay Nakrani 2023-09-16 12:36:49 -04:00
  • c577721a43
    Merge pull request #605 from jmorganca/mxyng/install.sh Michael Yang 2023-09-26 09:53:05 -07:00
  • 29c056ea39 ordered list of install locations Michael Yang 2023-09-26 09:38:11 -07:00
  • 9fc3bba9cf do no unload nouveau driver Michael Yang 2023-09-26 09:36:54 -07:00
  • 1c6c0fee82
    Added ollama gui interface Twan L 2023-09-26 09:04:45 -07:00
  • 7774ed4ae6
    Update README.md for linux + cleanup (#601) Michael Chiang 2023-09-25 23:44:53 -07:00
  • 1042276f52
    Update README.md Jeffrey Morgan 2023-09-25 23:44:01 -07:00
  • 0a8908683a
    Update README.md Jeffrey Morgan 2023-09-25 23:42:44 -07:00
  • a0416f57a0
    Update README.md Michael Chiang 2023-09-25 23:39:57 -07:00
  • a09b02e540
    Update README.md Michael Chiang 2023-09-25 23:32:01 -07:00
  • ef20cb82ff
    Update README.md Michael Chiang 2023-09-25 23:31:12 -07:00
  • e56a3a2487
    Update README.md Michael Chiang 2023-09-25 23:29:56 -07:00
  • 11f920f209
    Merge pull request #599 from jmorganca/mxyng/install.sh Michael Yang 2023-09-25 18:24:13 -07:00
  • 6e6b655956 update install.sh Michael Yang 2023-09-25 18:09:34 -07:00
  • 110ae89a6c
    Merge pull request #596 from jmorganca/mxyng/install.sh Michael Yang 2023-09-25 17:59:13 -07:00
  • 5e388f931e check cuda installed before installing Michael Yang 2023-09-25 17:56:43 -07:00
  • d5ad41dd7b fix path for wsl user Michael Yang 2023-09-25 17:56:25 -07:00
  • d294a11bc9 start service on exit instead of immediately Michael Yang 2023-09-25 16:11:21 -07:00
  • 93d887e4bc add painter message for exit Michael Yang 2023-09-25 16:30:14 -07:00
  • 3cfe97395c build slim, GPU-less docker image Michael Yang 2023-09-25 16:11:32 -07:00
  • 863e5a557e start service on exit instead of immediately Michael Yang 2023-09-25 16:11:21 -07:00
  • 5306b0269d
    Update linux.md v0.1.0 Jeffrey Morgan 2023-09-25 16:10:32 -07:00
  • 7de0c8345d
    Merge pull request #595 from jmorganca/mxyng/install.sh Michael Yang 2023-09-25 15:49:47 -07:00
  • 1b9dcab3ab ignore systemctl is-system-running exit code Michael Yang 2023-09-25 15:47:39 -07:00
  • 86279f4ae3
    unbound max num gpu layers (#591) Bruce MacDonald 2023-09-25 23:36:46 +01:00
  • b934bf23e6
    exit on unknown distro (#594) Michael Yang 2023-09-25 15:30:58 -07:00
  • 6fc2743934 Update llama.go Bruce MacDonald 2023-09-25 18:25:52 -04:00
  • acd9caffb7
    Update llm/llama.go Bruce MacDonald 2023-09-25 18:25:20 -04:00
  • fb783ea350
    Update llm/llama.go Bruce MacDonald 2023-09-25 18:25:08 -04:00
  • eb21c73aac exit on unknown distro Michael Yang 2023-09-25 15:19:48 -07:00
  • 51b110eb42 type casting fix Bruce MacDonald 2023-09-25 18:20:39 -04:00
  • 65422078b5 Update llama.go Bruce MacDonald 2023-09-25 18:17:26 -04:00
  • daee1cc361 Update llama.go Bruce MacDonald 2023-09-25 18:17:04 -04:00
  • 9cc4b879ad return int64 from numlayers Bruce MacDonald 2023-09-25 18:15:19 -04:00
  • d72c5ffb2b Update llama.go Bruce MacDonald 2023-09-25 18:13:25 -04:00
  • 22473b8618 Update gguf.go Bruce MacDonald 2023-09-25 18:04:46 -04:00