"docs/vscode:/vscode.git/clone" did not exist on "d11b5068dd74de6694cea0cce350bc86eb2ba5b2"
Unverified Commit f9fb33ef authored by Baizhou Zhang's avatar Baizhou Zhang Committed by GitHub
Browse files

Add 8-GPU Test for Deepseek-V3 (#5691)


Co-authored-by: default avatarLianmin Zheng <lianminzheng@gmail.com>
parent a38f6932
...@@ -102,7 +102,7 @@ jobs: ...@@ -102,7 +102,7 @@ jobs:
bash scripts/ci_install_dependency.sh bash scripts/ci_install_dependency.sh
- name: Run test - name: Run test
timeout-minutes: 30 timeout-minutes: 40
run: | run: |
cd test/srt cd test/srt
python3 run_suite.py --suite per-commit-8-gpu python3 run_suite.py --suite per-commit-8-gpu
......
...@@ -93,6 +93,7 @@ suites = { ...@@ -93,6 +93,7 @@ suites = {
], ],
"per-commit-8-gpu": [ "per-commit-8-gpu": [
TestFile("test_local_attn.py", 250), TestFile("test_local_attn.py", 250),
TestFile("test_full_deepseek_v3.py", 250),
], ],
"nightly": [ "nightly": [
TestFile("test_nightly_gsm8k_eval.py"), TestFile("test_nightly_gsm8k_eval.py"),
......
import unittest
from types import SimpleNamespace
import requests
import torch
from sglang.srt.utils import kill_process_tree
from sglang.test.few_shot_gsm8k import run_eval as run_eval_few_shot_gsm8k
from sglang.test.test_utils import (
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
DEFAULT_URL_FOR_TEST,
CustomTestCase,
is_in_ci,
popen_launch_server,
run_bench_one_batch,
write_github_step_summary,
)
FULL_DEEPSEEK_V3_MODEL_PATH = "deepseek-ai/DeepSeek-V3-0324"
class TestDeepseekV3(CustomTestCase):
@classmethod
def setUpClass(cls):
cls.model = FULL_DEEPSEEK_V3_MODEL_PATH
cls.base_url = DEFAULT_URL_FOR_TEST
other_args = ["--trust-remote-code", "--tp", "8"]
cls.process = popen_launch_server(
cls.model,
cls.base_url,
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
other_args=other_args,
)
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
def test_gsm8k(self):
args = SimpleNamespace(
num_shots=8,
data_path=None,
num_questions=1400,
parallel=1400,
max_new_tokens=512,
host="http://127.0.0.1",
port=int(self.base_url.split(":")[-1]),
)
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
self.assertGreater(metrics["accuracy"], 0.94)
class TestBenchOneBatch(CustomTestCase):
def test_bs1(self):
output_throughput = run_bench_one_batch(
FULL_DEEPSEEK_V3_MODEL_PATH,
["--trust-remote-code", "--tp", "8", "--cuda-graph-max-bs", "2"],
)
print(f"output_throughput : {output_throughput:.2f} token/s")
if is_in_ci():
write_github_step_summary(
f"### test_bs1\n"
f"output_throughput : {output_throughput:.2f} token/s\n"
)
self.assertGreater(output_throughput, 70)
class TestDeepseekV3MTP(CustomTestCase):
@classmethod
def setUpClass(cls):
cls.model = FULL_DEEPSEEK_V3_MODEL_PATH
cls.base_url = DEFAULT_URL_FOR_TEST
other_args = [
"--tp",
"8",
"--trust-remote-code",
"--speculative-algorithm",
"EAGLE",
"--speculative-draft",
"lmsys/DeepSeek-V3-0324-NextN",
"--speculative-num-steps",
"5",
"--speculative-eagle-topk",
"4",
"--speculative-num-draft-tokens",
"8",
"--mem-fraction-static",
"0.6",
]
cls.process = popen_launch_server(
cls.model,
cls.base_url,
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
other_args=other_args,
)
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
def test_gsm8k(self):
requests.get(self.base_url + "/flush_cache")
args = SimpleNamespace(
num_shots=5,
data_path=None,
num_questions=200,
max_new_tokens=512,
parallel=128,
host="http://127.0.0.1",
port=int(self.base_url.split(":")[-1]),
)
metrics = run_eval_few_shot_gsm8k(args)
print(f"{metrics=}")
self.assertGreater(metrics["accuracy"], 0.94)
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["avg_spec_accept_length"]
print(f"{avg_spec_accept_length=}")
self.assertGreater(avg_spec_accept_length, 3.2)
if __name__ == "__main__":
unittest.main()
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment