* Update README.md

Important Statement Regarding Supported Models Session

* Fix memory leak in quantize_i2_s

* update the new technical report in readme

* Fixing compilation error for ARM64+TL1 settings: microsoft#74

* Update README.md

* Add GCC to compiler check

* Fix compiler errors on GCC

* Add `-fpermissive` if using GCC

* Update README.md acknowledgement section

* Fix llama-bench path error on Windows

* Update README.md

* Add support for ios platform

* add fc3 support

* updated submodule

* update submodule

* fix weird character issue

* Make the coverage table more readable with both dark and light theme

* more changes to support chat models

* add 10b model

* fix issues

* add changes on README

* change

* fix version requirement of transformers pypi package and model list for codegen

* fix readme issue and -cnv option issue

* Fix model name in setup_env.py

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* update README

* add support for bitnet2b_2501 model

* add conversion logic for new model

* update readme and setup script to support official BitNet b1.58 model  (microsoft#171)

* update readme and setup file for new model.

* update model file name

---------

Co-authored-by: Yan Xia <[email protected]>

* Update README.md

* Update README.md (microsoft#172)

add two FAQs for windows build requestions.

* Update README.md

* add third-party demo

* refine readme

* Fix model architecture name

* Add run_inference_server.py for Running llama.cpp Built-in Server (microsoft#204)

* Update CMakeLists.txt

I added a CMake option to compile the Llama.cpp server. This update allows us to easily build and deploy the server using BitNet

* Create run_inference_server.py

same as run_inference, but for use with llama.cpp's built in server, for some extra comfort

In particular:
- The build directory is determined based on whether the system is running on Windows or not.
- A list of arguments (`--model`, `-m` etc.) is created.
- The main argument list is parsed and passed to the `subprocess.run()` method to execute the system command.

* Fix build error with GCC by forcing Clang compiler in CMake on android/aarch64 (microsoft#242)

GCC does not recognize Clang-specific warning flags like
-Wunreachable-code-break and -Wunreachable-code-return, which are passed
by upstream submodules (e.g., ggml). This patch forces CMake to use Clang
via command-line arguments, avoiding the need to patch nested submodules.

This resolves compiler errors without modifying submodule source code.

* add falcon-e support

* Update README.md

* Init gpu branch

* refine readme for gpu kernel

* feat: add also base models

* update submodule

* Enable conversion from .safetensors checkpoints to gguf files

* Update readme for gpu kernels

---------

Co-authored-by: Shaoguang Mao <[email protected]>
Co-authored-by: Yury <[email protected]>
Co-authored-by: Andre Buryndin <[email protected]>
Co-authored-by: Eddie-Wang <[email protected]>
Co-authored-by: Shuming Ma <[email protected]>
Co-authored-by: Goran Jelic-Cizmek <[email protected]>
Co-authored-by: Yan Xia <[email protected]>
Co-authored-by: potassiummmm <[email protected]>
Co-authored-by: potassiummmm <[email protected]>
Co-authored-by: younesbelkada <[email protected]>
Co-authored-by: Luca Foppiano <[email protected]>
Co-authored-by: tsong-ms <[email protected]>
Co-authored-by: Yan Xia <[email protected]>
Co-authored-by: tsong <[email protected]>
Co-authored-by: junhuihe <[email protected]>
Co-authored-by: Junhui He <[email protected]>
Co-authored-by: Benjamin Wegener <[email protected]>
Co-authored-by: Younes Belkada <[email protected]>
Co-authored-by: Junhui He <[email protected]>
Co-authored-by: younesbelkada <[email protected]>