[CI/Build] Reduce LoRA model tests. (#42266)

Signed-off-by: Jee Jee Li <jeejeelee@inferact.ai>
This commit is contained in:
Jee Jee Li
2026-05-11 14:49:08 +08:00
committed by GitHub
parent 581b5e9afc
commit 05d610e5cd
11 changed files with 50 additions and 31 deletions
+9
View File
@@ -1,9 +1,12 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import pytest
import vllm
import vllm.config
from vllm.lora.request import LoRARequest
from vllm.platforms import current_platform
from ..utils import create_new_process_for_each_test, multi_gpu_test
@@ -50,6 +53,9 @@ def do_sample(llm: vllm.LLM, lora_path: str, lora_id: int) -> list[str]:
return generated_texts
@pytest.mark.skipif(
current_platform.is_cuda_alike(), reason="Skipping to avoid redundant model tests"
)
@create_new_process_for_each_test()
def test_chatglm3_lora(chatglm3_lora_files):
llm = vllm.LLM(
@@ -70,6 +76,9 @@ def test_chatglm3_lora(chatglm3_lora_files):
assert output2[i] == EXPECTED_LORA_OUTPUT[i]
@pytest.mark.skipif(
current_platform.is_cuda_alike(), reason="Skipping to avoid redundant model tests"
)
@multi_gpu_test(num_gpus=4)
def test_chatglm3_lora_tp4(chatglm3_lora_files):
llm = vllm.LLM(