3 people 3d1f5e74ea Fix GPT-QModel compat and deprecate AutoGPTQ (#2426) ...

* remove references to deprecated exllama v1 kernel

* remove max_input_length from gptq api

* remove exllamav2-specific gptq docs and tests

* rename gptqmodel references to GPTQ-Model in docs

* restore gptq serialization test coverage

* remove autogptq support from gptq integration

* fix gptq formatting

* test real gptq post-init path

* update version

* fix tests/gptq/test_quantization.py

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

* Apply suggestion from @IlyasMoutawwakil

---------

Signed-off-by: ZX-ModelCloud <[email protected]>
Co-authored-by: ZX-ModelCloud <[email protected]>
Co-authored-by: Ilyas Moutawwakil <[email protected]>

2026-06-19 17:35:12 +02:00

..

cli

Misc v2 preparations (#2365)

2025-10-06 16:19:52 +02:00

common

Clean up workflows (#2231)

2025-04-19 13:09:28 +02:00

exporters/common

fix(exporters): handle read-only config in sentence-transformers >=5 (#2439)

2026-06-09 10:45:32 +02:00

fx

Misc v2 preparations (#2365)

2025-10-06 16:19:52 +02:00

gptq

Fix GPT-QModel compat and deprecate AutoGPTQ (#2426)

2026-06-19 17:35:12 +02:00

pipelines

Transformers v5 (#2408)

2026-03-13 16:41:20 +01:00

utils

fix tests preprocessor loading (#2445)

2026-06-03 16:15:24 +02:00

run_doctest.sh

Remove BetterTransformer (#2305)

2025-07-31 19:57:55 +02:00