Merge pull request #10 from arthw/fix_yaml

fix format

Merge pull request #8 from arthw/fix_q4_1

fix ut fault of Q4_1, Q5..

Merge pull request #7 from arthw/cherry_pick_20250224

Cherry pick 20250224

Merge pull request #6 from arthw/cherry-1220

Cherry 1220

Merge pull request #5 from arthw/cherry-1118

Cherry 1118

fix error

ggml-backend : fix async copy from CPU (ggml-org#8897)

* ggml-backend : fix async copy from CPU

* cuda : more reliable async copy, fix stream used when the devices are the same

[SYCL] Fixing wrong VDR iq4nl value (ggml-org#8812)

Merge pull request #2 from arthw/refactor_dev

Refactor device management and usage api

llama : add support for llama 3.1 rope scaling factors (ggml-org#8676)

* Add llama 3.1 rope scaling factors to llama conversion and inference

This commit generates the rope factors on conversion and adds them to the resulting model as a tensor. At inference time, these factors are passed to the `ggml_rope_ext` rope oepration, improving results for context windows above 8192

* Update convert_hf_to_gguf.py

Co-authored-by: compilade <[email protected]>

* address comments

* address comments

* Update src/llama.cpp

Co-authored-by: compilade <[email protected]>

* Update convert_hf_to_gguf.py

Co-authored-by: compilade <[email protected]>

---------

Co-authored-by: compilade <[email protected]>