[Feature] Support text encoder Disaggregation for H3 (#5885)

Signed-off-by: Canlin Guo <canlinguosdu@gmail.com>
Signed-off-by: lishunyang12 <lishunyang12@163.com>
Signed-off-by: yuanwu <yuan.wu@intel.com>
Co-authored-by: lishunyang12 <lishunyang12@163.com>
Co-authored-by: yuanwu <yuan.wu@intel.com>
This commit is contained in:
authored and GitHub committed 2026-08-28 14:27:05 +08:00
1 parent ed837e350e
commit b6a4baa659
61 files changed
+2952 -366

No files matched your search

@@ -47,7 +47,13 @@ def _build_test_app(
engine_client.abort = mocker.AsyncMock()
engine_client.submit_interaction_async = mocker.AsyncMock()
engine_client.default_sampling_params_list = [OmniDiffusionSamplingParams()]
engine_client.stage_configs = [SimpleNamespace(stage_type="diffusion")]
engine_client.stage_configs = [
SimpleNamespace(
stage_type="diffusion",
final_output=True,
final_output_type="video",
)
]
engine_client.generate = mock_generate
class FakeStreamingVideoEncoder:
@@ -72,10 +78,6 @@ def _build_test_app(
"vllm_omni.entrypoints.openai.serving_video_output_stream.create_streaming_video_encoder",
encoder_factory,
)
mocker.patch(
"vllm_omni.entrypoints.openai.serving_video_output_stream.get_stage_type",
return_value="diffusion",
)
mocker.patch(
"vllm_omni.entrypoints.openai.serving_video_output_stream.build_stage_sampling_params_list",
return_value=[OmniDiffusionSamplingParams()],
@@ -398,7 +400,13 @@ def _build_handler_for_async_tests(
engine_client.abort = mocker.AsyncMock()
engine_client.submit_interaction_async = mocker.AsyncMock()
engine_client.default_sampling_params_list = [OmniDiffusionSamplingParams()]
engine_client.stage_configs = [SimpleNamespace(stage_type="diffusion")]
engine_client.stage_configs = [
SimpleNamespace(
stage_type="diffusion",
final_output=True,
final_output_type="video",
)
]
engine_client.generate = mock_generate
chunks = streaming_chunks or [(b"mp4-chunk-0", True)]
@@ -421,10 +429,6 @@ def _build_handler_for_async_tests(
"vllm_omni.entrypoints.openai.serving_video_output_stream.create_streaming_video_encoder",
encoder_factory,
)
mocker.patch(
"vllm_omni.entrypoints.openai.serving_video_output_stream.get_stage_type",
return_value="diffusion",
)
mocker.patch(
"vllm_omni.entrypoints.openai.serving_video_output_stream.build_stage_sampling_params_list",
return_value=[OmniDiffusionSamplingParams()],