Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: ModelCloud/GPTQModel
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v1.7.2
Choose a base ref
...
head repository: ModelCloud/GPTQModel
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: v1.7.3
Choose a head ref
  • 18 commits
  • 24 files changed
  • 7 contributors

Commits on Jan 19, 2025

  1. start v1.7.3-dev cycle (#1104) 

    * start v1.7.3-dev cycle

* Update README.md
    @Qubitium
    Qubitium authored Jan 19, 2025
    Configuration menu
    Copy the full SHA
    2846042 View commit details
    Browse the repository at this point in the history

Commits on Jan 20, 2025

  1. [CI] do release on Xeon5 (#1105)

    @CSY-ModelCloud
    CSY-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    040099b View commit details
    Browse the repository at this point in the history

  2. [CI] add nvidia runtime args for unit tests (#1108) 

    * [CI] test runtime args

* [CI] pass gpu args

* [CI[ run nvidia-smi

* [CI] remove nvidia-smi

* [CI] assert gpu id is valid
    @CSY-ModelCloud
    CSY-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    cf88039 View commit details
    Browse the repository at this point in the history

  3. Add util.tensor_parameters() (#1107) 

    * add tensor_parameters()

* cleanup

* cleanup

* Update auto.py

* Update auto.py

* Update auto.py

* use math.ceil()

* Update auto.py

* Update auto.py

* cleanup

* cleanup

* cleanup

* add license

---------

Co-authored-by: Qubitium-ModelCloud <[email protected]>
    @ZX-ModelCloud @Qubitium
    ZX-ModelCloud and Qubitium authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    97d136a View commit details
    Browse the repository at this point in the history

  4. add require_dtype (#1109)

    LRL-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    e0ad9eb View commit details
    Browse the repository at this point in the history

  5. [MODEL] Add Telechat2 (China Telecom) (#1106) 

    * add support for telechat2

* Update telechat2.py

* Update auto.py

---------

Co-authored-by: xiayongqiang <[email protected]>
Co-authored-by: LRL-ModelCloud <[email protected]>
Co-authored-by: Qubitium-ModelCloud <[email protected]>
    @1096125073 @Qubitium
    4 people authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    23603f6 View commit details
    Browse the repository at this point in the history

  6. update news (#1110)

    @Qubitium
    Qubitium authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    5890cce View commit details
    Browse the repository at this point in the history

  7. [FIX] Filter weight-sharing tensors when save (#1112) 

    * Filter weight-sharing tensors when save

* add comment
    @ZX-ModelCloud
    ZX-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    442a0db View commit details
    Browse the repository at this point in the history

  8. Add telechat test (#1111) 

    * add TestTeleChat_2

* Update README.md

* format
    LRL-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    6229cce View commit details
    Browse the repository at this point in the history

  9. fix convert_gptq_to_mlx_weights (#1113)

    LRL-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    900746b View commit details
    Browse the repository at this point in the history

  10. [CI] fix assert (#1114)

    @CSY-ModelCloud
    CSY-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    49c9246 View commit details
    Browse the repository at this point in the history

  11. add test_parameter_count.py (#1115) 

    * add test_parameter_count.py

* cleanup

* cleanup
    @ZX-ModelCloud
    ZX-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    cf5f1b1 View commit details
    Browse the repository at this point in the history

  12. Add gpqa eval task (#1117) 

    * lm_eval add gpqa task

* add test code
    @CL-ModelCloud
    CL-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    2b3f5eb View commit details
    Browse the repository at this point in the history

  13. Call tied_weights() after load_checkpoint_in_model() to have the weig… 

    …hts tied correctly. (#1119)
    @ZX-ModelCloud
    ZX-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    bfae134 View commit details
    Browse the repository at this point in the history

  14. add phimoe support (#1118) 

    * add Phi

* add PhiMoE

* fix no layers_node

* use EXPERT_INDEX_PLACEHOLDER

* add phi moe test
    @CSY-ModelCloud
    CSY-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    fbafe07 View commit details
    Browse the repository at this point in the history

  15. [FIX] failed unit test (#1120) 

    * use local path

* The code for filter weight-sharing tensors refers to transformers v4.38.2 instead of v4.48.0
    @ZX-ModelCloud
    ZX-ModelCloud authored Jan 20, 2025
    Configuration menu
    Copy the full SHA
    d4c8c72 View commit details
    Browse the repository at this point in the history

Commits on Jan 21, 2025

  1. prepare for v1.7.3 release (#1121) 

    * prepare for v1.7.3 release

* Update README.md
    @Qubitium
    Qubitium authored Jan 21, 2025
    Configuration menu
    Copy the full SHA
    76d6aa3 View commit details
    Browse the repository at this point in the history

  2. Update README.md (#1122)

    @Qubitium
    Qubitium authored Jan 21, 2025
    Configuration menu
    Copy the full SHA
    5c1a7e8 View commit details
    Browse the repository at this point in the history
Loading