Gahow Wang
445e491123
third_party/vllm/ now tracked in git for direct patch management.
Based on vLLM v0.18.1 release with one patch applied:
vllm/v1/core/sched/scheduler.py:
Replace fatal assert with graceful skip when KV transfer callback
arrives for an already-aborted request during PD disaggregated serving.
Future vLLM modifications should be made directly in third_party/vllm/
and committed normally. The patches/ directory is kept as documentation
of what changed from upstream.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
33 lines
997 B
Python
33 lines
997 B
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
"""Test model set-up and inference for quantized HF models supported
|
|
on the AutoRound.
|
|
|
|
Validating the configuration and printing results for manual checking.
|
|
|
|
Run `pytest tests/quantization/test_auto_round.py`.
|
|
"""
|
|
|
|
import pytest
|
|
|
|
from vllm.platforms import current_platform
|
|
|
|
MODELS = [
|
|
"OPEA/Qwen2.5-0.5B-Instruct-int4-sym-inc", ##auto_round:auto_gptq
|
|
"Intel/Qwen2-0.5B-Instruct-int4-sym-AutoRound", ##auto_round:auto_awq
|
|
]
|
|
|
|
|
|
@pytest.mark.skipif(
|
|
not current_platform.is_cpu()
|
|
and not current_platform.is_xpu()
|
|
and not current_platform.is_cuda(),
|
|
reason="only supports CPU/XPU/CUDA backend.",
|
|
)
|
|
@pytest.mark.parametrize("model", MODELS)
|
|
def test_auto_round(vllm_runner, model):
|
|
with vllm_runner(model, enforce_eager=True) as llm:
|
|
output = llm.generate_greedy(["The capital of France is"], max_tokens=8)
|
|
assert output
|
|
print(f"{output[0][1]}")
|