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

[Misc] Clean up useless patch (#3320)

Browse Source 
### What this PR does / why we need it?
1. clean up v0.10.2 support in ut and e2e test
2. remove v0.11.0 period job, we're at v0.11.0 now.
3. remove uesless patch for deepseek v3.2. They have been done in vLLM
already.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?


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

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-10-09 14:07:26 +08:00
committed by GitHub
parent a43e2f61e1
commit 1c5b302f0d
10 changed files with 29 additions and 412 deletions
Download Patch File Download Diff File Expand all files Collapse all files

70
tests/e2e/singlecard/test_guided_decoding.py
View File

@@ -22,15 +22,8 @@ from typing import Any, Dict
import jsonschema
import pytest
import regex as re
from vllm_ascend.utils import vllm_version_is
if vllm_version_is("0.10.2"):
from vllm.sampling_params import GuidedDecodingParams, SamplingParams
else:
from vllm.sampling_params import SamplingParams, StructuredOutputsParams
from vllm.outputs import RequestOutput
from vllm.sampling_params import SamplingParams, StructuredOutputsParams
from tests.e2e.conftest import VllmRunner
@@ -91,27 +84,16 @@ def sample_json_schema():
def test_guided_json_completion(guided_decoding_backend: str,
sample_json_schema):
runner_kwargs: Dict[str, Any] = {}
if vllm_version_is("0.10.2"):
sampling_params = SamplingParams(
temperature=1.0,
max_tokens=500,
guided_decoding=GuidedDecodingParams(json=sample_json_schema))
runner_kwargs = {
"seed": 0,
"guided_decoding_backend": guided_decoding_backend,
}
else:
sampling_params = SamplingParams(
temperature=1.0,
max_tokens=500,
structured_outputs=StructuredOutputsParams(
json=sample_json_schema))
runner_kwargs = {
"seed": 0,
"structured_outputs_config": {
"backend": guided_decoding_backend
},
}
sampling_params = SamplingParams(
temperature=1.0,
max_tokens=500,
structured_outputs=StructuredOutputsParams(json=sample_json_schema))
runner_kwargs = {
"seed": 0,
"structured_outputs_config": {
"backend": guided_decoding_backend
},
}
with VllmRunner(MODEL_NAME, **runner_kwargs) as vllm_model:
prompts = [
f"Give an example JSON for an employee profile "
@@ -141,26 +123,16 @@ def test_guided_regex(guided_decoding_backend: str, sample_regex):
if guided_decoding_backend == "outlines":
pytest.skip("Outlines doesn't support regex-based guided decoding.")
runner_kwargs: Dict[str, Any] = {}
if vllm_version_is("0.10.2"):
sampling_params = SamplingParams(
temperature=0.8,
top_p=0.95,
guided_decoding=GuidedDecodingParams(regex=sample_regex))
runner_kwargs = {
"seed": 0,
"guided_decoding_backend": guided_decoding_backend,
}
else:
sampling_params = SamplingParams(
temperature=0.8,
top_p=0.95,
structured_outputs=StructuredOutputsParams(regex=sample_regex))
runner_kwargs = {
"seed": 0,
"structured_outputs_config": {
"backend": guided_decoding_backend
},
}
sampling_params = SamplingParams(
temperature=0.8,
top_p=0.95,
structured_outputs=StructuredOutputsParams(regex=sample_regex))
runner_kwargs = {
"seed": 0,
"structured_outputs_config": {
"backend": guided_decoding_backend
},
}
with VllmRunner(MODEL_NAME, **runner_kwargs) as vllm_model:
prompts = [