diff --git a/test/srt/layers/attention/mamba/test_causal_conv1d.py b/test/registered/layers/mamba/test_causal_conv1d.py similarity index 99% rename from test/srt/layers/attention/mamba/test_causal_conv1d.py rename to test/registered/layers/mamba/test_causal_conv1d.py index dd1a9a25f..a94b9d45b 100644 --- a/test/srt/layers/attention/mamba/test_causal_conv1d.py +++ b/test/registered/layers/mamba/test_causal_conv1d.py @@ -1,3 +1,7 @@ +from sglang.test.ci.ci_register import register_cuda_ci + +register_cuda_ci(est_time=25, suite="stage-b-test-small-1-gpu") + # Adapted from https://github.com/vllm-project/vllm/blob/main/tests/kernels/mamba/test_causal_conv1d.py diff --git a/test/srt/layers/attention/mamba/test_mamba2_mixer.py b/test/registered/layers/mamba/test_mamba2_mixer.py similarity index 97% rename from test/srt/layers/attention/mamba/test_mamba2_mixer.py rename to test/registered/layers/mamba/test_mamba2_mixer.py index 2252db653..1018a03b1 100644 --- a/test/srt/layers/attention/mamba/test_mamba2_mixer.py +++ b/test/registered/layers/mamba/test_mamba2_mixer.py @@ -1,3 +1,7 @@ +from sglang.test.ci.ci_register import register_cuda_ci + +register_cuda_ci(est_time=50, suite="stage-b-test-large-2-gpu") + # Adapted from https://github.com/vllm-project/vllm/blob/2c58742dff8613a3bd7496f2008ce927e18d38d1/tests/kernels/mamba/test_mamba_mixer2.py diff --git a/test/srt/layers/attention/mamba/test_mamba_ssm.py b/test/registered/layers/mamba/test_mamba_ssm.py similarity index 98% rename from test/srt/layers/attention/mamba/test_mamba_ssm.py rename to test/registered/layers/mamba/test_mamba_ssm.py index 4a2c9a8e2..264859974 100644 --- a/test/srt/layers/attention/mamba/test_mamba_ssm.py +++ b/test/registered/layers/mamba/test_mamba_ssm.py @@ -1,3 +1,7 @@ +from sglang.test.ci.ci_register import register_cuda_ci + +register_cuda_ci(est_time=7, suite="stage-b-test-small-1-gpu") + # Adapted from https://github.com/vllm-project/vllm/blob/633f943e30a4444d890d26b81850f7217736f840/tests/kernels/mamba/test_mamba_ssm_ssd.py diff --git a/test/srt/layers/attention/mamba/test_mamba_ssm_ssd.py b/test/registered/layers/mamba/test_mamba_ssm_ssd.py similarity index 99% rename from test/srt/layers/attention/mamba/test_mamba_ssm_ssd.py rename to test/registered/layers/mamba/test_mamba_ssm_ssd.py index 10a7f3f80..b06cb5528 100644 --- a/test/srt/layers/attention/mamba/test_mamba_ssm_ssd.py +++ b/test/registered/layers/mamba/test_mamba_ssm_ssd.py @@ -1,3 +1,7 @@ +from sglang.test.ci.ci_register import register_cuda_ci + +register_cuda_ci(est_time=13, suite="stage-b-test-small-1-gpu") + # Adapted from https://github.com/vllm-project/vllm/blob/633f943e30a4444d890d26b81850f7217736f840/tests/kernels/mamba/test_mamba_ssm_ssd.py diff --git a/test/srt/run_suite.py b/test/srt/run_suite.py index d46ed0be0..9c0f8e06c 100644 --- a/test/srt/run_suite.py +++ b/test/srt/run_suite.py @@ -12,9 +12,6 @@ suites = { TestFile("debug_utils/test_tensor_dump_forward_hook.py", 9), TestFile("hicache/test_hicache_storage.py", 96), TestFile("hicache/test_hicache_variants.py", 368), - TestFile("layers/attention/mamba/test_causal_conv1d.py", 25), - TestFile("layers/attention/mamba/test_mamba_ssm.py", 7), - TestFile("layers/attention/mamba/test_mamba_ssm_ssd.py", 13), TestFile("openai_server/basic/test_openai_embedding.py", 70), TestFile("openai_server/basic/test_openai_server.py", 184), TestFile("openai_server/basic/test_protocol.py", 3), @@ -92,7 +89,6 @@ suites = { TestFile("hicache/test_hicache_storage_3fs_backend.py", 200), TestFile("hicache/test_hicache_storage_file_backend.py", 200), TestFile("hicache/test_hicache_storage_mooncake_backend.py", 300), - TestFile("layers/attention/mamba/test_mamba2_mixer.py", 50), TestFile("models/test_kimi_linear_models.py", 90), TestFile("models/test_nvidia_nemotron_nano_v2.py", 132), TestFile("rl/test_update_weights_from_distributed.py", 103),