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

[CI] Enable linux-aarch64-a2 (64GB) and tp2 * 2 max-parallel to speed up CI (#2065)

Browse Source 
### What this PR does / why we need it?
Currently our workflow run time takes about 3 hours in total, which
seriously affects the developer experience, so it is urgent to have a
optimization, after this pr, It is expected that the running time of the
full CI can be shortened to 1h40min.

- Enable linux-aarch64-a2 (64GB) to replace linux-arm64-npu (32GB)
- Change TP4 ---> TP2 * 2 max-parallel
- Move DeepSeek-V2-Lite-W8A8 to single card test

### Does this PR introduce _any_ user-facing change?
No


- vLLM version: v0.10.0
- vLLM main:
a2480251ec

---------

Signed-off-by: wangli <wangli858794774@gmail.com>
This commit is contained in:
Li Wang
2025-07-29 18:59:05 +08:00
committed by GitHub
parent ca8007f584
commit f60bb474f9
14 changed files with 75 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
tests/e2e/multicard/test_offline_inference_distributed.py
View File

@@ -42,7 +42,7 @@ def test_models_distributed_QwQ():
with VllmRunner(
"Qwen/QwQ-32B",
dtype=dtype,
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
) as vllm_model:
vllm_model.generate_greedy(example_prompts, max_tokens)
@@ -57,7 +57,7 @@ def test_models_distributed_DeepSeek_multistream_moe():
with VllmRunner(
"vllm-ascend/DeepSeek-V3-Pruning",
dtype=dtype,
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
additional_config={
"torchair_graph_config": {
@@ -82,7 +82,7 @@ def test_models_distributed_DeepSeek_dbo():
with VllmRunner(
"deepseek-ai/DeepSeek-V2-Lite",
dtype=dtype,
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
) as vllm_model:
model_arch = 'DeepseekV2ForCausalLM'
@@ -106,7 +106,7 @@ def test_models_distributed_DeepSeekV3_dbo():
with VllmRunner(
"vllm-ascend/DeepSeek-V3-Pruning",
dtype=dtype,
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
) as vllm_model:
model_arch = 'DeepseekV3ForCausalLM'
@@ -118,24 +118,6 @@ def test_models_distributed_DeepSeekV3_dbo():
vllm_model.generate(example_prompts, sampling_params)
@pytest.mark.skip(reason="Due to OOM,waiting for 1311pr to merge in")
def test_models_distributed_DeepSeek_W8A8():
example_prompts = [
"Hello, my name is",
]
max_tokens = 5
with VllmRunner(
snapshot_download("vllm-ascend/DeepSeek-V2-Lite-W8A8"),
max_model_len=8192,
enforce_eager=True,
dtype="auto",
tensor_parallel_size=4,
quantization="ascend",
) as vllm_model:
vllm_model.generate_greedy(example_prompts, max_tokens)
def test_models_distributed_pangu():
example_prompts = [
"Hello, my name is",
@@ -147,7 +129,7 @@ def test_models_distributed_pangu():
max_model_len=8192,
enforce_eager=True,
dtype="auto",
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
) as vllm_model:
vllm_model.generate_greedy(example_prompts, max_tokens)
@@ -169,7 +151,7 @@ def test_models_distributed_topk() -> None:
with VllmRunner(
"deepseek-ai/DeepSeek-V2-Lite",
dtype=dtype,
tensor_parallel_size=4,
tensor_parallel_size=2,
distributed_executor_backend="mp",
) as vllm_model:
vllm_model.generate(example_prompts, sampling_params)
@@ -186,7 +168,7 @@ def test_models_distributed_Qwen3_W8A8():
max_model_len=8192,
enforce_eager=True,
dtype="auto",
tensor_parallel_size=4,
tensor_parallel_size=2,
quantization="ascend",
) as vllm_model:
vllm_model.generate_greedy(example_prompts, max_tokens)