Hexiang Wang 0ad52517a1 Revert "Refactor quantization layer name mapping to leverage vLLM built-in mappers" (#7237) ...

Reverts vllm-project/vllm-ascend#7050, which breaks kimi-k2.5 and
qwen-omin.
- vLLM version: v0.17.0
- vLLM main:
4034c3d32e

2026-03-14 00:05:54 +08:00

..

methods

[Feature] support aclgraph for model runner v2 (#7110)

2026-03-13 09:11:46 +08:00

__init__.py

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #7) (#6023)

2026-02-06 14:56:53 +08:00

compressed_tensors_config.py

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #7) (#6023)

2026-02-06 14:56:53 +08:00

method_adapters.py

[bugdix] The problem that the w4a8 weight fails to be loaded when the EP is not enabled is resolved. (#7090)

2026-03-10 16:57:05 +08:00

modelslim_config.py

Revert "Refactor quantization layer name mapping to leverage vLLM built-in mappers" (#7237)

2026-03-14 00:05:54 +08:00

quant_parser.py

[misc] move mxfp_compat into device to decouple from quantization init chain (#6918)

2026-03-02 18:17:01 +08:00

utils.py

[Feature][Quant] Reapply auto-detect quantization format and support remote model ID (#7111)

2026-03-13 22:53:25 +08:00