EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Model][1/N] Delete deepseek v2/v3 modeling codes. (#3189)

Browse Source 
This PR deletes model codes of deepseek_v2 and deepseek_v3 to reuse the
model file from vLLM.

vLLM Ascend now uses custom ops register way instead of model file
hard-coding.

- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: whx-sjtu <2952154980@qq.com>
This commit is contained in:
whx
2025-10-20 15:31:34 +08:00
committed by GitHub
parent 918ded9155
commit f8b52fe950
13 changed files with 143 additions and 843 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
tests/ut/models/conftest.py
View File

@@ -90,13 +90,7 @@ def mock_distributed():
mock_vllm_config.scheduler_config = Mock(max_num_seqs=256)
mock_vllm_config.model_config = Mock(max_model_len=2048, quant_config=None)
with patch("vllm_ascend.models.deepseek_v2.get_tensor_model_parallel_rank", return_value=0), \
patch("vllm_ascend.models.deepseek_v2.get_tensor_model_parallel_world_size", return_value=1), \
patch("vllm_ascend.models.deepseek_v2.get_tp_group", return_value=tp_group), \
patch("vllm_ascend.models.deepseek_v2.get_pp_group", return_value=pp_group), \
patch("vllm_ascend.models.deepseek_v2.get_pp_group",
return_value=Mock(is_first_rank=False, is_last_rank=False)), \
patch("vllm_ascend.ops.common_fused_moe.get_current_vllm_config", return_value=mock_vllm_config), \
with patch("vllm_ascend.ops.common_fused_moe.get_current_vllm_config", return_value=mock_vllm_config), \
patch("vllm_ascend.ops.moe.token_dispatcher.torch.distributed.get_rank", return_value=0), \
patch("vllm_ascend.ops.moe.token_dispatcher.get_ascend_soc_version", return_value=None), \
patch.dict("vllm.distributed.parallel_state.__dict__", _TP=tp_group, _EP=ep_group, _DP=dp_group,
@@ -104,11 +98,3 @@ def mock_distributed():
patch.dict("vllm_ascend.distributed.parallel_state.__dict__", _MC2=ep_group), \
patch("torch.npu.current_device", return_value=0):
yield
@pytest.fixture
def mock_forward_context():
forward_context = Mock(in_profile_run=False, with_prefill=False)
with patch("vllm_ascend.models.deepseek_v2.get_forward_context",
return_value=forward_context):
yield