mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-16 10:26:07 +08:00
17 lines
637 B
Python
17 lines
637 B
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
import pytest
|
|
import torch
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"model_path",
|
|
[("nm-testing/SpeculatorLlama3-1-8B-Eagle3-converted-0717"),
|
|
("nm-testing/SpeculatorLlama3-1-8B-Eagle3-converted-0717-quantized")])
|
|
def test_llama(vllm_runner, example_prompts, model_path):
|
|
with vllm_runner(model_path, dtype=torch.bfloat16) as vllm_model:
|
|
vllm_outputs = vllm_model.generate_greedy(example_prompts,
|
|
max_tokens=20)
|
|
print(vllm_outputs)
|
|
assert vllm_outputs
|