Angazenn 0037d461b0 [Feature] Adapt to new quantization_config (#38) ...

This pr mainly focuses on: 
1. adapting to new quantization_config generated from msmodelslim.
2. removing unnecessary imports.
3. disable warning when loading BasevLLMParameter

---------

Signed-off-by: angazenn <zengyanjia@huawei.com>
Co-authored-by: angazenn <zengyanjia@huawei.com>

2025-02-17 22:25:21 +08:00

..

quantization

[Feature] Adapt to new quantization_config (#38)

2025-02-17 22:25:21 +08:00

__init__.py

[Core]Add Ascend Quantize (#7)

2025-02-11 09:47:28 +08:00

conftest.py

[Core] Init vllm-ascend (#3)

2025-02-05 10:53:12 +08:00

model_utils.py

[Core] Init vllm-ascend (#3)

2025-02-05 10:53:12 +08:00

test_offline_inference.py

[Core] Init vllm-ascend (#3)

2025-02-05 10:53:12 +08:00