[diffusion] refactor: refactor pipeline folders (#13253)

python/sglang/multimodal_gen/configs/pipeline_configs/__init__.py

@@ -1,16 +1,16 @@
 # Copied and adapted from: https://github.com/hao-ai-lab/FastVideo
 
-from sglang.multimodal_gen.configs.pipelines.base import (
+from sglang.multimodal_gen.configs.pipeline_configs.base import (
     PipelineConfig,
     SlidingTileAttnConfig,
 )
-from sglang.multimodal_gen.configs.pipelines.flux import FluxPipelineConfig
-from sglang.multimodal_gen.configs.pipelines.hunyuan import (
+from sglang.multimodal_gen.configs.pipeline_configs.flux import FluxPipelineConfig
+from sglang.multimodal_gen.configs.pipeline_configs.hunyuan import (
     FastHunyuanConfig,
     HunyuanConfig,
 )
-from sglang.multimodal_gen.configs.pipelines.stepvideo import StepVideoT2VConfig
-from sglang.multimodal_gen.configs.pipelines.wan import (
+from sglang.multimodal_gen.configs.pipeline_configs.stepvideo import StepVideoT2VConfig
+from sglang.multimodal_gen.configs.pipeline_configs.wan import (
     SelfForcingWanT2V480PConfig,
     WanI2V480PConfig,
     WanI2V720PConfig,

python/sglang/multimodal_gen/configs/pipeline_configs/flux.py

@@ -13,12 +13,12 @@ from sglang.multimodal_gen.configs.models.encoders import (
     T5Config,
 )
 from sglang.multimodal_gen.configs.models.vaes.flux import FluxVAEConfig
-from sglang.multimodal_gen.configs.pipelines.base import (
+from sglang.multimodal_gen.configs.pipeline_configs.base import (
     ModelTaskType,
     PipelineConfig,
     preprocess_text,
 )
-from sglang.multimodal_gen.configs.pipelines.hunyuan import (
+from sglang.multimodal_gen.configs.pipeline_configs.hunyuan import (
     clip_postprocess_text,
     clip_preprocess_text,
 )

python/sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py

@@ -15,7 +15,7 @@ from sglang.multimodal_gen.configs.models.encoders import (
     LlamaConfig,
 )
 from sglang.multimodal_gen.configs.models.vaes import HunyuanVAEConfig
-from sglang.multimodal_gen.configs.pipelines.base import PipelineConfig
+from sglang.multimodal_gen.configs.pipeline_configs import PipelineConfig
 
 PROMPT_TEMPLATE_ENCODE_VIDEO = (
     "<|start_header_id|>system<|end_header_id|>\n\nDescribe the video by detailing the following aspects: "

python/sglang/multimodal_gen/configs/pipeline_configs/qwen_image.py

@@ -9,7 +9,10 @@ from sglang.multimodal_gen.configs.models import DiTConfig, EncoderConfig, VAECo
 from sglang.multimodal_gen.configs.models.dits.qwenimage import QwenImageDitConfig
 from sglang.multimodal_gen.configs.models.encoders.qwen_image import Qwen2_5VLConfig
 from sglang.multimodal_gen.configs.models.vaes.qwenimage import QwenImageVAEConfig
-from sglang.multimodal_gen.configs.pipelines.base import ModelTaskType, PipelineConfig
+from sglang.multimodal_gen.configs.pipeline_configs.base import (
+    ModelTaskType,
+    PipelineConfig,
+)
 from sglang.multimodal_gen.utils import calculate_dimensions
 
 

python/sglang/multimodal_gen/configs/pipeline_configs/stepvideo.py

@@ -6,7 +6,7 @@ from dataclasses import dataclass, field
 from sglang.multimodal_gen.configs.models import DiTConfig, VAEConfig
 from sglang.multimodal_gen.configs.models.dits import StepVideoConfig
 from sglang.multimodal_gen.configs.models.vaes import StepVideoVAEConfig
-from sglang.multimodal_gen.configs.pipelines.base import PipelineConfig
+from sglang.multimodal_gen.configs.pipeline_configs.base import PipelineConfig
 
 
 @dataclass

python/sglang/multimodal_gen/configs/pipeline_configs/wan.py

@@ -14,7 +14,10 @@ from sglang.multimodal_gen.configs.models.encoders import (
     T5Config,
 )
 from sglang.multimodal_gen.configs.models.vaes import WanVAEConfig
-from sglang.multimodal_gen.configs.pipelines.base import ModelTaskType, PipelineConfig
+from sglang.multimodal_gen.configs.pipeline_configs.base import (
+    ModelTaskType,
+    PipelineConfig,
+)
 from sglang.multimodal_gen.runtime.utils.logging_utils import init_logger
 
 logger = init_logger(__name__)