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: v5.6.0
Choose a base ref
...
head repository: ModelCloud/GPTQModel
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: v5.6.2
Choose a head ref
  • 14 commits
  • 31 files changed
  • 6 contributors

Commits on Dec 10, 2025

  1. FIX some models not honoring model.config.use_cache by force pass use… 

    …_cache=false (#2246)

* disable cache

* add comments

* Update looper_helpers.py

* Add comment about use_cache handling in models

Added a comment regarding the use_cache property in models.

* Fix typo in TODO comment about use_cache property

* Fix typo in TODO comment about use_cache property

---------

Co-authored-by: Qubitium-ModelCloud <[email protected]>
    @LRL2-ModelCloud @Qubitium
    LRL2-ModelCloud and Qubitium authored Dec 10, 2025
    Configuration menu
    Copy the full SHA
    95063e2 View commit details
    Browse the repository at this point in the history

  2. Update version.py (#2247)

    @Qubitium
    Qubitium authored Dec 10, 2025
    Configuration menu
    Copy the full SHA
    23458a7 View commit details
    Browse the repository at this point in the history

  3. [FIX] JIT Pytorch extension pack_cpu_ext stall (#2248) 

    * add safe_load_cpp_ext()

Signed-off-by: ZX-ModelCloud <[email protected]>

* fix safe_load_cpp_ext() lock

Signed-off-by: ZX-ModelCloud <[email protected]>

* Revert "fix safe_load_cpp_ext() lock"

This reverts commit 34f6373.

* Create build_directory if it does not exist

Signed-off-by: ZX-ModelCloud <[email protected]>

* format

Signed-off-by: ZX-ModelCloud <[email protected]>

* Update cpp.py

---------

Signed-off-by: ZX-ModelCloud <[email protected]>
Co-authored-by: Qubitium-ModelCloud <[email protected]>
    @ZX-ModelCloud @Qubitium
    ZX-ModelCloud and Qubitium authored Dec 10, 2025
    Configuration menu
    Copy the full SHA
    35b228f View commit details
    Browse the repository at this point in the history

  4. Update README.md (#2250) 

    Fixed a very minor typo in the recent release notes.
    @davedgd
    davedgd authored Dec 10, 2025
    Configuration menu
    Copy the full SHA
    5bf5c1a View commit details
    Browse the repository at this point in the history

Commits on Dec 11, 2025

  1. Refractor Kernel External Dependency Validation (#2249) 

    * add validate_once()

* add cache_validate_once

* bitblas and machete use cache_validate_once

* mod exllama kernel

* mod exllama2

* fixed machete

* fix maxsize

* fix method name

* fix

* cleanup

* mod marlin_awq

* mod qqq

* cleanup

* cleanup

* cleanup

* cleanup

* fix import

* fix import

* cleanup

* format code

* make validate_one return a tuple of bool and optional[exception]
    @LRL2-ModelCloud
    LRL2-ModelCloud authored Dec 11, 2025
    Configuration menu
    Copy the full SHA
    f26c09d View commit details
    Browse the repository at this point in the history

  2. [CI] add torch 2.9.1 (#2254)

    @CSY-ModelCloud
    CSY-ModelCloud authored Dec 11, 2025
    Configuration menu
    Copy the full SHA
    8dddd6a View commit details
    Browse the repository at this point in the history

Commits on Dec 12, 2025

  1. fix INT3 BUG (#2258)

    @KingdalfGoodman
    KingdalfGoodman authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    4443148 View commit details
    Browse the repository at this point in the history

  2. Update license declaration in pyproject.toml (#2259) 

    * Update license declaration in pyproject.toml

* Remove license-files from pyproject.toml

* Include licenses directory in MANIFEST.in

* Downgrade version from 5.7.0 to 5.6.2
    @CSY-ModelCloud
    CSY-ModelCloud authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    a4dad1a View commit details
    Browse the repository at this point in the history

  3. Modify setup (#2260) 

    * device-smi/logbar depend udpate

* log when we cannot auto-download pre-compiled whls

* pypcre depend update
    @Qubitium
    Qubitium authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    4ec0370 View commit details
    Browse the repository at this point in the history

  4. Add release notes for version 5.6.2 (#2261) 

    * Add release notes for version 5.6.2

Updated latest news section with recent release notes.

* Update latest news section in README.md
    @Qubitium
    Qubitium authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    e1bbaae View commit details
    Browse the repository at this point in the history

  5. change model (#2262)

    @LRL2-ModelCloud
    LRL2-ModelCloud authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    e534536 View commit details
    Browse the repository at this point in the history

  6. [CI] mount dateset dir to /monster/data/model/dataset (#2263)

    @CSY-ModelCloud
    CSY-ModelCloud authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    a72941c View commit details
    Browse the repository at this point in the history

  7. fix parsing args (#2264) 

    * fix parsing args

* fix parsing args

* [CI] build eora for test
    @CSY-ModelCloud
    CSY-ModelCloud authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    48c557e View commit details
    Browse the repository at this point in the history

  8. Support llm-awq (#2252) 

    * add TestInferenceOnly::test_inference_quantized_by_llm_awq

Signed-off-by: ZX-ModelCloud <[email protected]>

* AwqGEMVFastQuantLinear adapted for llm-awq

Signed-off-by: ZX-ModelCloud <[email protected]>

* check AWQ_PACKING_BACKEND_FIELD

Signed-off-by: ZX-ModelCloud <[email protected]>

* AwqGEMVFastQuantLinear supports two zeros name

Signed-off-by: ZX-ModelCloud <[email protected]>

* add BACKEND.LLM_AWQ and FORMAT.LLM_AWQ

Signed-off-by: ZX-ModelCloud <[email protected]>

* removed BACKEND.LLM_AWQ

Signed-off-by: ZX-ModelCloud <[email protected]>

* cleanup

Signed-off-by: ZX-ModelCloud <[email protected]>

---------

Signed-off-by: ZX-ModelCloud <[email protected]>
    @ZX-ModelCloud
    ZX-ModelCloud authored Dec 12, 2025
    Configuration menu
    Copy the full SHA
    d97478f View commit details
    Browse the repository at this point in the history
Loading