Commits

Commit history

1. Commits on May 27, 2023

   • Use async routes

     abetlen

     committed
     80066f0View commit details

     Copy full SHA for 80066f0

     View code at this pointBrowse repository at this point

   • Import unnused import

     abetlen

     committed
     c2b59a5View commit details

     Copy full SHA for c2b59a5

     View code at this pointBrowse repository at this point
2. Commits on May 19, 2023

   • feat: make embedding support list of string as input

     Show description for e783f1c
     chabotsi

     committed
     e783f1cView commit details

     Copy full SHA for e783f1c

     View code at this pointBrowse repository at this point

   • Bugfix: Stop sequences can be strings

     abetlen

     committed
     a8cd169View commit details

     Copy full SHA for a8cd169

     View code at this pointBrowse repository at this point

   • Use server sent events function for streaming completion

     abetlen

     committed
     dc39cc0View commit details

     Copy full SHA for dc39cc0

     View code at this pointBrowse repository at this point
3. Commits on May 16, 2023

   • Add model_alias option to override model_path in completions. Closes #39

     abetlen

     committed
     a335292View commit details

     Copy full SHA for a335292

     View code at this pointBrowse repository at this point
4. Commits on May 14, 2023

   • Update llama.cpp

     abetlen

     committed
     cdf5976View commit details

     Copy full SHA for cdf5976

     View code at this pointBrowse repository at this point
5. Commits on May 12, 2023

   • Only support generating one prompt at a time.

     abetlen

     committed
     8740ddcView commit details

     Copy full SHA for 8740ddc

     View code at this pointBrowse repository at this point

   • Revert "llama_cpp server: prompt is a string". Closes #187

     Show description for 8895b90
     abetlen

     committed
     8895b90View commit details

     Copy full SHA for 8895b90

     View code at this pointBrowse repository at this point
6. Commits on May 9, 2023

   • llama_cpp server: document presence_penalty and frequency_penalty, mark as supported

     Stonelinks

     committed
     02e8a01View commit details

     Copy full SHA for 02e8a01

     View code at this pointBrowse repository at this point
7. Commits on May 8, 2023

   • Fix: default repeat_penalty

     abetlen

     committed
     82d138fView commit details

     Copy full SHA for 82d138f

     View code at this pointBrowse repository at this point

   • Set repeat_penalty to 0 by default

     abetlen

     committed
     0d751a6View commit details

     Copy full SHA for 0d751a6

     View code at this pointBrowse repository at this point

   • Add openai frequency and presence penalty parameters. Closes #169

     abetlen

     committed
     65d9cc0View commit details

     Copy full SHA for 65d9cc0

     View code at this pointBrowse repository at this point

   • Bugfix for models endpoint

     abetlen

     committed
     a0b61eaView commit details

     Copy full SHA for a0b61ea

     View code at this pointBrowse repository at this point
8. Commits on May 7, 2023

   • Added cache size to settins object.

     abetlen

     committed
     14da46fView commit details

     Copy full SHA for 14da46f

     View code at this pointBrowse repository at this point

   • Add verbose flag to server

     abetlen

     committed
     627811eView commit details

     Copy full SHA for 627811e

     View code at this pointBrowse repository at this point

   • Fix mlock_supported and mmap_supported return type

     abetlen

     committed
     3fbda71View commit details

     Copy full SHA for 3fbda71

     View code at this pointBrowse repository at this point

   • Update cpu_count

     abetlen

     committed
     5a3413eView commit details

     Copy full SHA for 5a3413e

     View code at this pointBrowse repository at this point

   • Update settings fields and defaults

     abetlen

     committed
     1a00e45View commit details

     Copy full SHA for 1a00e45

     View code at this pointBrowse repository at this point

   • Revert "llama_cpp server: delete some ignored / unused parameters"

     Show description for 8675397
     abetlen

     committed
     8675397View commit details

     Copy full SHA for 8675397

     View code at this pointBrowse repository at this point

   • Revert "llama_cpp server: mark model as required"

     Show description for c382d8f
     abetlen

     committed
     c382d8fView commit details

     Copy full SHA for c382d8f

     View code at this pointBrowse repository at this point
9. Commits on May 2, 2023

   • llama_cpp server: prompt is a string

     Show description for b9098b0
     Stonelinks

     committed
     b9098b0View commit details

     Copy full SHA for b9098b0

     View code at this pointBrowse repository at this point

   • Merge branch 'main' into better-server-params-and-fields

     abetlen

     authored
     7ab08b8View commit details

     Copy full SHA for 7ab08b8

     View code at this pointBrowse repository at this point

   • Refactor server to use factory

     abetlen

     committed
     9eafc4cView commit details

     Copy full SHA for 9eafc4c

     View code at this pointBrowse repository at this point
10. Commits on May 1, 2023

    • llama_cpp server: fix to ChatCompletionRequestMessage

      Show description for dbbfc4b
      Stonelinks

      committed
      dbbfc4bView commit details

      Copy full SHA for dbbfc4b

      View code at this pointBrowse repository at this point

    • llama_cpp server: fields for the embedding endpoint

      Stonelinks

      committed
      fa2a61eView commit details

      Copy full SHA for fa2a61e

      View code at this pointBrowse repository at this point

    • llama_cpp server: define fields for chat completions

      Show description for 8dcbf65
      Stonelinks

      committed
      8dcbf65View commit details

      Copy full SHA for 8dcbf65

      View code at this pointBrowse repository at this point

    • llama_cpp server: add some more information to fields for completions

      Stonelinks

      committed
      978b6daView commit details

      Copy full SHA for 978b6da

      View code at this pointBrowse repository at this point

    • llama_cpp server: add missing top_k param to CreateChatCompletionRequest

      Show description for a5aa6c1
      Stonelinks

      committed
      a5aa6c1View commit details

      Copy full SHA for a5aa6c1

      View code at this pointBrowse repository at this point

    • llama_cpp server: move logprobs to supported

      Show description for 1e42913
      Stonelinks

      committed
      1e42913View commit details

      Copy full SHA for 1e42913

      View code at this pointBrowse repository at this point

    • llama_cpp server: delete some ignored / unused parameters

      Show description for b47b954
      Stonelinks

      committed
      b47b954View commit details

      Copy full SHA for b47b954

      View code at this pointBrowse repository at this point

    • llama_cpp server: mark model as required

      Show description for e40fcb0
      Stonelinks

      committed
      e40fcb0View commit details

      Copy full SHA for e40fcb0

      View code at this pointBrowse repository at this point

    • Fix import error

      abetlen

      committed
      9ff9cddView commit details

      Copy full SHA for 9ff9cdd

      View code at this pointBrowse repository at this point
11. Commits on Apr 29, 2023

    • llama_cpp server: slight refactor to init_llama function

      Show description for efe8e6f
      Stonelinks

      committed
      efe8e6fView commit details

      Copy full SHA for efe8e6f

      View code at this pointBrowse repository at this point

    • tests: simple test for server module

      Stonelinks

      committed
      6d8db9dView commit details

      Copy full SHA for 6d8db9d

      View code at this pointBrowse repository at this point