chore: update CI test est_time values (#22565)

Co-authored-by: Cheng Wan <chwan@rice.edu>
Co-authored-by: sglang-bot <sglang-bot@users.noreply.github.com>
Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>

test/registered/models/test_compressed_tensors_models.py

@@ -13,7 +13,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cuda_ci(est_time=42, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=63, suite="stage-b-test-1-gpu-large")
 register_amd_ci(est_time=42, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/models/test_generation_models.py

@@ -1,7 +1,7 @@
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
 # Generation model tests (CUDA only)
-register_cuda_ci(est_time=103, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=124, suite="stage-b-test-1-gpu-large")
 register_amd_ci(est_time=106, suite="stage-b-test-1-gpu-small-amd")
 
 # Copyright 2023-2024 SGLang Team

test/registered/models/test_kimi_linear_models.py

@@ -11,7 +11,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cuda_ci(est_time=180, suite="stage-b-test-2-gpu-large")
+register_cuda_ci(est_time=168, suite="stage-b-test-2-gpu-large")
 
 
 class TestKimiLinear(CustomTestCase):

test/registered/models/test_nvidia_nemotron_3_nano.py

@@ -4,7 +4,7 @@ from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.kits.lm_eval_kit import LMEvalMixin
 from sglang.test.server_fixtures.default_fixture import DefaultServerBase
 
-register_cuda_ci(est_time=540, suite="stage-b-test-2-gpu-large")
+register_cuda_ci(est_time=564, suite="stage-b-test-2-gpu-large")
 
 NEMOTRON_3_NANO_THINKING_ARGS = [
     "--trust-remote-code",

test/registered/models/test_nvidia_nemotron_nano_v2.py

@@ -5,7 +5,7 @@ from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.kits.eval_accuracy_kit import GSM8KMixin
 from sglang.test.server_fixtures.default_fixture import DefaultServerBase
 
-register_cuda_ci(est_time=240, suite="stage-b-test-2-gpu-large")
+register_cuda_ci(est_time=214, suite="stage-b-test-2-gpu-large")
 
 
 class TestNvidiaNemotronNanoV2BF16(GSM8KMixin, DefaultServerBase):

test/registered/models/test_nvidia_nemotron_nano_v2_vl.py

@@ -10,7 +10,7 @@ from sglang.test.server_fixtures.mmmu_fixture import MMMUServerBase
 # GSM8k + MMMU evaluation
 
 
-register_cuda_ci(est_time=214, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=206, suite="stage-b-test-1-gpu-large")
 
 MODEL = "nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16"
 

test/registered/models/test_qwen_models.py

@@ -13,7 +13,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cuda_ci(est_time=90, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=109, suite="stage-b-test-1-gpu-small")
 register_amd_ci(est_time=130, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/models/test_transformers_backend_eval.py

@@ -7,7 +7,7 @@ from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.few_shot_gsm8k import run_eval
 from sglang.test.server_fixtures.default_fixture import DefaultServerBase
 
-register_cuda_ci(est_time=40, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=44, suite="stage-b-test-1-gpu-small")
 
 
 class TestTransformersBackendEval(DefaultServerBase):

test/registered/models/test_transformers_models.py

@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cuda_ci(est_time=260, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=244, suite="stage-b-test-1-gpu-small")
 register_amd_ci(est_time=320, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/models/test_vlm_models.py

@@ -13,7 +13,7 @@ from sglang.test.test_utils import is_in_ci
 # VLM (Vision Language Model) tests
 
 
-register_cuda_ci(est_time=228, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=176, suite="stage-b-test-1-gpu-large")
 register_amd_ci(est_time=850, suite="stage-b-test-1-gpu-small-amd-nondeterministic")
 
 _is_hip = is_hip()