* [CI] make name short

* [CI] rename to public server

* [CI] rename to task list

* [CI] rename to public server

* [CI] remove -zen4 suffix

* [CI] remove tag

* [CI] rename to runner

* [CI] fix env.GPU not exists

* [CI] don't run at last step

* move lm-eval to utils

* rename

* add hint to install lm-eval

* update unit tests

* add ipex bench code.

* update batch

* use cli arg

* print all arg values before benchmark starts

* add quantized model support

* cleanup

* Rename prompts.json to prompts.jsonl

---------

Co-authored-by: LRL-ModelCloud <[email protected]>
Co-authored-by: Qubitium-ModelCloud <[email protected]>

* add ProgressBar

* add time countdown

* replace with new progress bar

* remove suffix text

* should't print \n at last

* replace all tqdm usages

* remove tqdm in req

* rename func

* fix current not +1

* fix current not +1

* remove name

* fix iteration = 0

* add hymba support

* fix kv_last_layer

* fix layer cache

* padding outfeatures

* don't quantize "mamba.x_proj.0" and "mamba.dt_proj.0". Otherwise the quant model will output empty.

* add TODO

* change RREADME.md

---------

Co-authored-by: LRL-ModelCloud <[email protected]>
Co-authored-by: ZX-ModelCloud <[email protected]>

* Update README.md

* Update README.md

* Update README.md

…r of cores to control the number of threads used by OpenMP (#671)

Co-authored-by: LRL-ModelCloud <[email protected]>

Co-authored-by: LRL-ModelCloud <[email protected]>

* prep for 1.3.0 release

* Update README.md