anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

[CI] Reducing mithril-h100-pool in ready&merge CI, optimize ready CI duration (#4354)

Browse Source 
Signed-off-by: [Your Name] <your.email@example.com>
Signed-off-by: wangyu <410167048@qq.com>
Signed-off-by: wangyu <53896905+yenuo26@users.noreply.github.com>
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Signed-off-by: [Your Name] <[Your Email]>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
wangyuandHongsheng Liu authored and GitHub committed 2026-06-29 05:17:57 +00:00
1 parent e4026a50ad
commit 7619ed6b82
84 files changed
+1718 -2428
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

tests/entrypoints/openai_api/test_serving_speech_batch.py
+9 -9
View File
@@ -125,7 +125,7 @@ class TestSpeechBatchE2E:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_single_item(self, omni_server) -> None:
"""Batch with a single item works correctly."""
@@ -147,7 +147,7 @@ class TestSpeechBatchE2E:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_per_item_voice_override(self, omni_server) -> None:
"""Per-item voice overrides the batch-level default."""
@@ -175,7 +175,7 @@ class TestSpeechBatchE2E:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_multiple_languages(self, omni_server) -> None:
"""Batch items with different languages per item."""
@@ -200,7 +200,7 @@ class TestSpeechBatchE2E:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_whisper_transcription(self, omni_server) -> None:
"""Whisper transcription of batch output matches input text."""
@@ -235,7 +235,7 @@ class TestSpeechBatchE2E:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_result_indices_ordered(self, omni_server) -> None:
"""Result indices match the input item order."""
@@ -262,7 +262,7 @@ class TestSpeechBatchValidation:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_empty_items_rejected(self, omni_server) -> None:
"""Empty items list returns a 4xx error."""
@@ -276,7 +276,7 @@ class TestSpeechBatchValidation:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", default_server_params, indirect=True)
def test_batch_exceeds_max_items(self, omni_server) -> None:
"""Batch exceeding 32 items returns 400 error."""
@@ -334,7 +334,7 @@ class TestSpeechBatchSize2:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", batch2_server_params, indirect=True)
def test_batch2_produces_valid_audio(self, omni_server) -> None:
"""Batch of 2 items with batched engine produces valid audio."""
@@ -371,7 +371,7 @@ class TestSpeechBatchSize2:
@pytest.mark.advanced_model
@pytest.mark.tts
@hardware_test(res={"cuda": "H100"}, num_cards=1)
@hardware_test(res={"cuda": "L4"}, num_cards=1)
@pytest.mark.parametrize("omni_server", batch2_server_params, indirect=True)
def test_batch2_vs_sequential_timing(self, omni_server) -> None:
"""Compare batch-of-2 vs 2 sequential single requests.