Skip to content
This repository was archived by the owner on Jul 4, 2025. It is now read-only.

Conversation

@hiento09
Copy link

@hiento09 hiento09 commented Jun 23, 2024

  • Sync 0.10.0 from remote
  • Support llama3

* cpp-tiktoken integration init

* remove cpp-tiktoken folder

* tiktoken integration init

* move fmt to third-party dependency

* remove unnecessary option in cortex CMakeLists

* remove unnecessary submodule

* make pcre2 as third-party lib

* fix CI bug build in window

* fix CI bug in linux

* link directory pcre2

* Fix CI bug in linux 'pcre2.h' not found

* Fix bug build static third party lib

* Integate Llama3 successfully

* refactor with model type

* remove unnecessary comment

* format and code convention

* fix logic of mistral when preparing input

* Add remove end_of_text and end_of_turn token feature for llama3

* fix: patch pcre2 CMakeLists.txt file

* fix: do not add fPIC to cpp-tiktoken cmake file

* fix: suppress warnings

* fix: add CMAKE_C_FLAGS

* fix: CI bug build for window in cpp-tiktoken

* fix: CI bug for window fmt build fail without utf-8 flag

* fix: CI bug for window stop using fmt

* fix: build bugs for window - using latest pcre2 third party

* fix: build bugs for window - pcre2 static runtime on

* fix CI bug for window build pcre2

* rename link lib to pcre2-8-static in windows

* fix: use pcre2 dynamic lib for windows

* fix: pack tensorrt_llm_nvrtc_wrapper.dll

* test: only windows

* fix: disable setLoggerFinder for msvc

* fix: cleanup cmake files

* fix: remove fmt

* fix: pack libtensorrt_llm_nvrtc_wrapper

* fix: patches windows

* fix: rm fmt

---------

Co-authored-by: sangjanai <sang@jan.ai>
@vansangpfiev vansangpfiev merged commit 3c01167 into rel Jul 25, 2024
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants