first pass

fastvideo/configs/pipelines/hunyuan15.py

@@ -9,7 +9,7 @@ import torch
 from fastvideo.configs.models import DiTConfig, EncoderConfig, VAEConfig
 from fastvideo.configs.models.dits import HunyuanVideo15Config
 from fastvideo.configs.models.encoders import (BaseEncoderOutput,
-                                               Qwen2_5_VLConfig, T5Config)
+                                               Qwen2_5_VLConfig, T5Config, SiglipVisionConfig)
 from fastvideo.configs.models.vaes import Hunyuan15VAEConfig
 from fastvideo.configs.pipelines.base import PipelineConfig
 
@@ -137,3 +137,13 @@ class Hunyuan15T2V720PConfig(Hunyuan15T2V480PConfig):
 
     # HunyuanConfig-specific parameters with defaults
     flow_shift: int = 9
+
+@dataclass
+class Hunyuan15I2V480PConfig(Hunyuan15T2V480PConfig):
+    image_encoder_config: EncoderConfig = field(
+        default_factory=SiglipVisionConfig)
+    image_encoder_precision: str = "bf16"
+
+    def __post_init__(self):
+        self.vae_config.load_encoder = True
+        self.vae_config.load_decoder = True

fastvideo/configs/pipelines/registry.py

@@ -8,7 +8,7 @@ from fastvideo.configs.pipelines.base import PipelineConfig
 from fastvideo.configs.pipelines.cosmos import CosmosConfig
 from fastvideo.configs.pipelines.cosmos2_5 import Cosmos25Config
 from fastvideo.configs.pipelines.hunyuan import FastHunyuanConfig, HunyuanConfig
-from fastvideo.configs.pipelines.hunyuan15 import Hunyuan15T2V480PConfig, Hunyuan15T2V720PConfig
+from fastvideo.configs.pipelines.hunyuan15 import Hunyuan15T2V480PConfig, Hunyuan15T2V720PConfig, Hunyuan15I2V480PConfig
 from fastvideo.configs.pipelines.hyworld import HYWorldConfig
 from fastvideo.configs.pipelines.ltx2 import LTX2T2VConfig
 from fastvideo.configs.pipelines.stepvideo import StepVideoT2VConfig
@@ -37,6 +37,8 @@ PIPE_NAME_TO_CONFIG: dict[str, type[PipelineConfig]] = {
     "hunyuanvideo-community/HunyuanVideo": HunyuanConfig,
     "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v":
     Hunyuan15T2V480PConfig,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v":
+    Hunyuan15I2V480PConfig,
     "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v":
     Hunyuan15T2V720PConfig,
     "FastVideo/HY-WorldPlay-Bidirectional-Diffusers": HYWorldConfig,

fastvideo/configs/sample/hunyuan15.py

@@ -17,8 +17,6 @@ class Hunyuan15_480P_SamplingParam(SamplingParam):
     guidance_scale: float = 6.0
     prompt_attention_mask: list = field(default_factory=list)
     negative_attention_mask: list = field(default_factory=list)
-    sigmas: list[float] | None = field(
-        default_factory=lambda: list(np.linspace(1.0, 0.0, 50 + 1)[:-1]))
 
     negative_prompt: str = ""
 

fastvideo/configs/sample/registry.py

@@ -46,6 +46,8 @@ SAMPLING_PARAM_REGISTRY: dict[str, Any] = {
     "hunyuanvideo-community/HunyuanVideo": HunyuanSamplingParam,
     "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_t2v":
     Hunyuan15_480P_SamplingParam,
+    "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-480p_i2v":
+    Hunyuan15_480P_SamplingParam,
     "hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v":
     Hunyuan15_720P_SamplingParam,
     "FastVideo/HY-WorldPlay-Bidirectional-Diffusers": HYWorld_SamplingParam,

fastvideo/pipelines/basic/hunyuan15/hunyuan15_i2v_pipeline.py

@@ -0,0 +1,80 @@
+# SPDX-License-Identifier: Apache-2.0
+"""
+HYWorld video diffusion pipeline implementation.
+
+This module contains an implementation of the HYWorld video diffusion pipeline
+using the modular pipeline architecture with HYWorld-specific denoising stage
+for chunk-based video generation with context frame selection.
+"""
+
+from fastvideo.fastvideo_args import FastVideoArgs
+from fastvideo.logger import init_logger
+from fastvideo.pipelines.composed_pipeline_base import ComposedPipelineBase
+from fastvideo.pipelines.stages import (
+    ConditioningStage, DecodingStage, DenoisingStage,
+    InputValidationStage, LatentPreparationStage, TextEncodingStage,
+    TimestepPreparationStage, HYWorldImageEncodingStage)
+
+logger = init_logger(__name__)
+
+
+class HunyuanVideo15ImageToVideoPipeline(ComposedPipelineBase):
+    """
+    HunyuanVideo15 image to video pipeline.
+
+    This pipeline implements image to video generation using the HunyuanVideo15 model.
+    """
+
+    # Include image_encoder and feature_extractor for I2V support with SigLIP
+    # Note: guider (ClassifierFreeGuidance) is not needed - FastVideo handles CFG differently
+    _required_config_modules = [
+        "text_encoder", "tokenizer", "vae", "transformer", "scheduler",
+        "text_encoder_2", "tokenizer_2", "image_encoder", "feature_extractor"
+    ]
+
+    def create_pipeline_stages(self, fastvideo_args: FastVideoArgs):
+        """Set up pipeline stages with HYWorld-specific denoising stage."""
+
+        self.add_stage(stage_name="input_validation_stage",
+                       stage=InputValidationStage())
+
+        self.add_stage(stage_name="prompt_encoding_stage_primary",
+                       stage=TextEncodingStage(
+                           text_encoders=[
+                               self.get_module("text_encoder"),
+                               self.get_module("text_encoder_2")
+                           ],
+                           tokenizers=[
+                               self.get_module("tokenizer"),
+                               self.get_module("tokenizer_2")
+                           ]))
+
+        self.add_stage(stage_name="conditioning_stage",
+                       stage=ConditioningStage())
+
+        self.add_stage(stage_name="timestep_preparation_stage",
+                       stage=TimestepPreparationStage(
+                           scheduler=self.get_module("scheduler")))
+
+        self.add_stage(stage_name="latent_preparation_stage",
+                       stage=LatentPreparationStage(
+                           scheduler=self.get_module("scheduler"),
+                           transformer=self.get_module("transformer")))
+
+        self.add_stage(stage_name="image_encoding_stage",
+                       stage=HYWorldImageEncodingStage(
+                           image_encoder=self.get_module("image_encoder"),
+                           image_processor=self.get_module("feature_extractor"),
+                           vae=self.get_module("vae")))
+
+        self.add_stage(stage_name="denoising_stage",
+                       stage=DenoisingStage(
+                           transformer=self.get_module("transformer"),
+                           scheduler=self.get_module("scheduler"),
+                           pipeline=self))
+
+        self.add_stage(stage_name="decoding_stage",
+                       stage=DecodingStage(vae=self.get_module("vae")))
+
+
+EntryClass = HunyuanVideo15ImageToVideoPipeline

fastvideo/pipelines/pipeline_registry.py

@@ -28,6 +28,7 @@ _PIPELINE_NAME_TO_ARCHITECTURE_NAME: dict[str, str] = {
     "StepVideoPipeline": "stepvideo",
     "HunyuanVideoPipeline": "hunyuan",
     "HunyuanVideo15Pipeline": "hunyuan15",
+    "HunyuanVideo15ImageToVideoPipeline": "hunyuan15",
     "HYWorldPipeline": "hyworld",
     "Cosmos2VideoToWorldPipeline": "cosmos",
     "Cosmos2_5Pipeline": "cosmos",