minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[CI] [NPU] npu ci use existing modelscope model (#17868)

Browse Source
This commit is contained in:
Even Zhou
2026-01-28 16:51:41 +08:00
committed by GitHub
parent c08b54a575
commit 6077de1237
9 changed files with 9 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/srt/ascend/test_ascend_compile_graph_tp1_bf16.py
+1 -1
View File
@@ -15,7 +15,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.84, "accuracy": 0.84,
"latency": 150, "latency": 150,
"output_throughput": 30, "output_throughput": 30,
test/srt/ascend/test_ascend_graph_tp1_bf16.py
+1 -1
View File
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.85, "accuracy": 0.85,
"latency": 150, "latency": 150,
"output_throughput": 30, "output_throughput": 30,
test/srt/ascend/test_ascend_graph_tp2_bf16.py
+1 -1
View File
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.85, "accuracy": 0.85,
"latency": 180, "latency": 180,
"output_throughput": 20, "output_throughput": 20,
test/srt/ascend/test_ascend_hicache_mha.py
+1 -1
View File
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.85, "accuracy": 0.85,
"latency": 150, "latency": 150,
"output_throughput": 30, "output_throughput": 30,
test/srt/ascend/test_ascend_piecewise_graph_prefill.py
+1 -1
View File
@@ -12,7 +12,7 @@ from sglang.test.test_utils import (
run_bench_one_batch, run_bench_one_batch,
) )
MODEL = "Qwen/Qwen2.5-7B-Instruct" MODEL = "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct"
GSM8K_EXP_ACCURACY = 0.84 GSM8K_EXP_ACCURACY = 0.84
EXP_PREFILL_LATENCY = 0.045 EXP_PREFILL_LATENCY = 0.045
TOKENS_TO_CAPTURE = [i for i in range(128, 4096, 128)] TOKENS_TO_CAPTURE = [i for i in range(128, 4096, 128)]
test/srt/ascend/test_ascend_sampling_backend.py
+1 -1
View File
@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
class TestAscendSamplingBackend(CustomTestCase): class TestAscendSamplingBackend(CustomTestCase):
@classmethod @classmethod
def setUpClass(cls): def setUpClass(cls):
cls.model = "Qwen/Qwen2.5-7B-Instruct" cls.model = "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct"
cls.base_url = DEFAULT_URL_FOR_TEST cls.base_url = DEFAULT_URL_FOR_TEST
cls.process = popen_launch_server( cls.process = popen_launch_server(
cls.model, cls.model,
test/srt/ascend/test_ascend_tp1_bf16.py
+1 -1
View File
@@ -12,7 +12,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.84, "accuracy": 0.84,
"latency": 150, "latency": 150,
"output_throughput": 30, "output_throughput": 30,
test/srt/ascend/test_ascend_tp2_bf16.py
+1 -1
View File
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.85, "accuracy": 0.85,
"latency": 180, "latency": 180,
"output_throughput": 20, "output_throughput": 20,
test/srt/ascend/test_ascend_tp2_fia_bf16.py
+1 -1
View File
@@ -13,7 +13,7 @@ from sglang.test.test_utils import (
) )
TEST_MODEL_MATRIX = { TEST_MODEL_MATRIX = {
"Qwen/Qwen2.5-7B-Instruct": { "/root/.cache/modelscope/hub/models/Qwen/Qwen2.5-7B-Instruct": {
"accuracy": 0.85, "accuracy": 0.85,
"latency": 180, "latency": 180,
"output_throughput": 20, "output_throughput": 20,