3f3c201243
Co-authored-by: kkHuang-amd <wunhuang@amd.com> Co-authored-by: YC Tseng <yctseng@amd.com>
104 lines
2.9 KiB
Python
104 lines
2.9 KiB
Python
import unittest
|
|
from types import SimpleNamespace
|
|
|
|
import requests
|
|
|
|
from sglang import Engine
|
|
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
|
|
|
|
register_cuda_ci(est_time=103, suite="stage-b-test-small-1-gpu")
|
|
register_amd_ci(est_time=230, suite="stage-b-test-small-1-gpu-amd")
|
|
from sglang.lang.chat_template import get_chat_template_by_model_path
|
|
from sglang.srt.utils import kill_process_tree
|
|
from sglang.test.run_eval import run_eval
|
|
from sglang.test.test_utils import (
|
|
DEFAULT_IMAGE_URL,
|
|
DEFAULT_MODEL_NAME_FOR_TEST,
|
|
DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST,
|
|
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
|
|
DEFAULT_URL_FOR_TEST,
|
|
CustomTestCase,
|
|
is_in_amd_ci,
|
|
popen_launch_server,
|
|
)
|
|
|
|
|
|
class TestTorchAO(CustomTestCase):
|
|
@classmethod
|
|
def setUpClass(cls):
|
|
cls.model = DEFAULT_MODEL_NAME_FOR_TEST
|
|
cls.base_url = DEFAULT_URL_FOR_TEST
|
|
cls.process = popen_launch_server(
|
|
cls.model,
|
|
cls.base_url,
|
|
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
|
|
other_args=["--torchao-config", "int4wo-128"],
|
|
)
|
|
|
|
@classmethod
|
|
def tearDownClass(cls):
|
|
kill_process_tree(cls.process.pid)
|
|
|
|
def test_mmlu(self):
|
|
args = SimpleNamespace(
|
|
base_url=self.base_url,
|
|
model=self.model,
|
|
eval_name="mmlu",
|
|
num_examples=64,
|
|
num_threads=32,
|
|
)
|
|
|
|
metrics = run_eval(args)
|
|
assert metrics["score"] >= 0.60
|
|
|
|
def run_decode(self, max_new_tokens):
|
|
response = requests.post(
|
|
self.base_url + "/generate",
|
|
json={
|
|
"text": "The capital of France is",
|
|
"sampling_params": {
|
|
"temperature": 0,
|
|
"max_new_tokens": max_new_tokens,
|
|
},
|
|
"ignore_eos": True,
|
|
},
|
|
)
|
|
return response.json()
|
|
|
|
def test_throughput(self):
|
|
import time
|
|
|
|
max_tokens = 256
|
|
|
|
tic = time.perf_counter()
|
|
res = self.run_decode(max_tokens)
|
|
tok = time.perf_counter()
|
|
print(res["text"])
|
|
throughput = max_tokens / (tok - tic)
|
|
print(f"Throughput: {throughput} tokens/s")
|
|
if is_in_amd_ci():
|
|
assert throughput >= 150
|
|
else:
|
|
assert throughput >= 210
|
|
|
|
|
|
class TestTorchAOForVLM(CustomTestCase):
|
|
def test_vlm_generate(self):
|
|
model_path = DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST
|
|
chat_template = get_chat_template_by_model_path(model_path)
|
|
text = f"{chat_template.image_token}What is in this picture? Answer: "
|
|
|
|
engine = Engine(
|
|
model_path=model_path,
|
|
max_total_tokens=512,
|
|
enable_multimodal=True,
|
|
torchao_config="fp8wo",
|
|
)
|
|
out = engine.generate([text], image_data=[DEFAULT_IMAGE_URL])
|
|
engine.shutdown()
|
|
self.assertGreater(len(out), 0)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|