Hugging Face's logo

dzmdd
/
Fast_deploy

Model card Files
xet
 Community
Fast_deploy / ComfyUI-MiniMaxH3-Context-Loop /ComfyUI-MAINodes /examples /motion_pipeline_ref2va_api.json
dzmdd's picture
dzmdd
Upload folder using huggingface_hub
e488ce7 verified
Raw History Blame Contribute Delete
11.2 kB
{
"300": {
"class_type": "PathchSageAttentionKJ",
"inputs": {
"model": [
"127",
0
],
"sage_attention": "auto",
"allow_compile": false
},
"_meta": {
"title": "PathchSageAttentionKJ"
}
},
"301": {
"class_type": "MiniMaxH3MemoryEfficientSageAttentionPatch",
"inputs": {
"model": [
"300",
0
]
},
"_meta": {
"title": "MiniMaxH3MemoryEfficientSageAttentionPatch"
}
},
"302": {
"class_type": "MiniMaxChunkFeedForward",
"inputs": {
"model": [
"301",
0
],
"chunks": 2,
"seq_threshold": 65536
},
"_meta": {
"title": "MiniMaxChunkFeedForward"
}
},
"303": {
"class_type": "LoraLoaderModelOnly",
"inputs": {
"model": [
"302",
0
],
"lora_name": "minimax_h3-lightx2v/minimax_h3_fl2v_turbo_4step_v1.0_768p_comfyui_bf16.safetensors",
"strength_model": 1.0
},
"_meta": {
"title": "LoraLoaderModelOnly"
}
},
"115": {
"class_type": "ResolutionSelector",
"inputs": {
"aspect_ratio": "1:1 (Square)",
"megapixels": 1.0,
"multiple": 32
},
"_meta": {
"title": "ResolutionSelector"
}
},
"119": {
"class_type": "VAELoader",
"inputs": {
"vae_name": "minimax_h3/minimax_h3_video_vae_fp16.safetensors"
},
"_meta": {
"title": "VAELoader"
}
},
"120": {
"class_type": "VAELoader",
"inputs": {
"vae_name": "minimax_h3/minimax_h3_audio_vae_fp32.safetensors"
},
"_meta": {
"title": "VAELoader"
}
},
"127": {
"class_type": "UNETLoader",
"inputs": {
"unet_name": "minimax_h3/minimax_h3_ref2va_pruned_int8_convrot.safetensors",
"weight_dtype": "default"
},
"_meta": {
"title": "UNETLoader"
}
},
"128": {
"class_type": "CLIPLoader",
"inputs": {
"clip_name": "minimax_h3/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
"type": "minimax",
"device": "default"
},
"_meta": {
"title": "CLIPLoader"
}
},
"133": {
"class_type": "PrimitiveFloat",
"inputs": {
"value": 4.0
},
"_meta": {
"title": "PrimitiveFloat"
}
},
"132": {
"class_type": "ComfyMathExpression",
"inputs": {
"expression": "max(5, round(a * 24)) + (5 - (max(5, round(a * 24)) % 17)) % 17",
"values.a": [
"133",
0
]
},
"_meta": {
"title": "ComfyMathExpression"
}
},
"234": {
"class_type": "LoadImage",
"inputs": {
"image": "example.png"
},
"_meta": {
"title": "LoadImage"
}
},
"129": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 20260901
},
"_meta": {
"title": "RandomNoise"
}
},
"131": {
"class_type": "MiniMaxH3ReferenceToVideo",
"inputs": {
"clip": [
"128",
0
],
"vae": [
"119",
0
],
"audio_vae": [
"120",
0
],
"prompt": "subject_definitions:\n<Subject 1> is the person shown in <Picture 1>; their face, hair and clothing are carried into the target video unchanged.\n\nsummary:\n[reference generation] The target video is a single continuous shot of <Subject 1> performing a fast turning movement, with their appearance taken from <Picture 1>.\n\nretention_analysis:\n<Subject 1> is fully preserved from <Picture 1> -- facial features, hair and clothing stay consistent in every frame. The pose in <Picture 1> is used only as the starting pose and is not held.\n\ndetailed_description:\n[Shot 1] Live-action, cinematic, a medium-wide shot frames <Subject 1> standing on open ground in even daylight. They step forward onto the left foot, swing both arms across the body and spin a full fast turn, their hair and clothing flaring outward with the rotation, then plant both feet and stop square to the camera, breathing quickly. The camera arcs left at fast speed around them through the turn, then settles.\n\noverall_soundscape: Footsteps scuffing and thudding on dry ground; the rush and snap of clothing turning through air; a light wind; a sharp exhale at the stop and quick breathing after it.\n\nnon_diegetic_music: N/A",
"width": [
"115",
0
],
"height": [
"115",
1
],
"length": [
"132",
1
],
"ref_image_size": "match",
"ref_images.ref_image_0": [
"234",
0
]
},
"_meta": {
"title": "MiniMaxH3ReferenceToVideo"
}
},
"124": {
"class_type": "BasicScheduler",
"inputs": {
"model": [
"302",
0
],
"scheduler": "linear_quadratic",
"steps": 12,
"denoise": 1.0
},
"_meta": {
"title": "BasicScheduler"
}
},
"126": {
"class_type": "BasicGuider",
"inputs": {
"model": [
"302",
0
],
"conditioning": [
"131",
0
]
},
"_meta": {
"title": "BasicGuider"
}
},
"135": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "gradient_estimation"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"125": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"noise": [
"129",
0
],
"guider": [
"126",
0
],
"sampler": [
"135",
0
],
"sigmas": [
"124",
0
],
"latent_image": [
"131",
1
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"122": {
"class_type": "VAEDecode",
"inputs": {
"samples": [
"125",
0
],
"vae": [
"119",
0
]
},
"_meta": {
"title": "VAEDecode"
}
},
"121": {
"class_type": "VAEDecodeAudio",
"inputs": {
"samples": [
"125",
0
],
"vae": [
"120",
0
]
},
"_meta": {
"title": "VAEDecodeAudio"
}
},
"130": {
"class_type": "CreateVideo",
"inputs": {
"images": [
"122",
0
],
"audio": [
"121",
0
],
"fps": 24,
"bit_depth": 8
},
"_meta": {
"title": "CreateVideo"
}
},
"92": {
"class_type": "SaveVideo",
"inputs": {
"video": [
"130",
0
],
"format": "auto",
"codec": "auto",
"filename_prefix": "video/mainodes_ref2va_baseline"
},
"_meta": {
"title": "SaveVideo"
}
},
"200": {
"class_type": "H3JerkOracle",
"inputs": {
"samples": [
"125",
0
],
"length": [
"132",
1
],
"q": 0.75,
"d_max": 4,
"ramp": true,
"preset": "balanced (default)",
"bridge": 8
},
"_meta": {
"title": "H3JerkOracle"
}
},
"201": {
"class_type": "H3TimeSmear",
"inputs": {
"images": [
"122",
0
],
"dilation": 4,
"hold_map": [
"200",
0
]
},
"_meta": {
"title": "H3TimeSmear"
}
},
"202": {
"class_type": "VAEEncode",
"inputs": {
"pixels": [
"201",
0
],
"vae": [
"119",
0
]
},
"_meta": {
"title": "VAEEncode"
}
},
"203": {
"class_type": "H3V2VInit",
"inputs": {
"samples": [
"202",
0
],
"audio_latent": [
"9102",
0
],
"audio_mode": "follow the original performance (0.5)",
"audio_strength": 0.5
},
"_meta": {
"title": "H3V2VInit"
}
},
"205": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 20260902
},
"_meta": {
"title": "RandomNoise"
}
},
"206": {
"class_type": "MiniMaxH3ReferenceToVideo",
"inputs": {
"clip": [
"128",
0
],
"vae": [
"119",
0
],
"audio_vae": [
"120",
0
],
"prompt": "subject_definitions:\n<Subject 1> is the person shown in <Picture 1>; their face, hair and clothing are carried into the target video unchanged.\n\nsummary:\n[reference generation] The target video is a single continuous shot of <Subject 1> performing a fast turning movement, with their appearance taken from <Picture 1>.\n\nretention_analysis:\n<Subject 1> is fully preserved from <Picture 1> -- facial features, hair and clothing stay consistent in every frame. The pose in <Picture 1> is used only as the starting pose and is not held.\n\ndetailed_description:\n[Shot 1] Live-action, cinematic, a medium-wide shot frames <Subject 1> standing on open ground in even daylight. They step forward onto the left foot, swing both arms across the body and spin a full fast turn, their hair and clothing flaring outward with the rotation, then plant both feet and stop square to the camera, breathing quickly. The camera arcs left at fast speed around them through the turn, then settles.\n\noverall_soundscape: Footsteps scuffing and thudding on dry ground; the rush and snap of clothing turning through air; a light wind; a sharp exhale at the stop and quick breathing after it.\n\nnon_diegetic_music: N/A",
"width": [
"115",
0
],
"height": [
"115",
1
],
"length": [
"132",
1
],
"ref_image_size": "match",
"ref_images.ref_image_0": [
"234",
0
]
},
"_meta": {
"title": "MiniMaxH3ReferenceToVideo"
}
},
"204": {
"class_type": "H3InjectSchedule",
"inputs": {
"model": [
"303",
0
],
"scheduler": "beta",
"total_steps": 6,
"inject": 0.7,
"preset": "faithful detail 0.50 (metric best)"
},
"_meta": {
"title": "H3InjectSchedule"
}
},
"207": {
"class_type": "BasicGuider",
"inputs": {
"model": [
"303",
0
],
"conditioning": [
"206",
0
]
},
"_meta": {
"title": "BasicGuider"
}
},
"136": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "gradient_estimation"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"208": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"noise": [
"205",
0
],
"guider": [
"207",
0
],
"sampler": [
"136",
0
],
"sigmas": [
"204",
0
],
"latent_image": [
"203",
0
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"209": {
"class_type": "VAEDecode",
"inputs": {
"samples": [
"208",
0
],
"vae": [
"119",
0
]
},
"_meta": {
"title": "VAEDecode"
}
},
"210": {
"class_type": "VAEDecodeAudio",
"inputs": {
"samples": [
"208",
0
],
"vae": [
"120",
0
]
},
"_meta": {
"title": "VAEDecodeAudio"
}
},
"213": {
"class_type": "H3ExactRecover",
"inputs": {
"images": [
"209",
0
],
"hold_map": [
"201",
1
]
},
"_meta": {
"title": "H3ExactRecover"
}
},
"219": {
"class_type": "H3AudioRecover",
"inputs": {
"audio": [
"210",
0
],
"hold_map": [
"201",
1
],
"fps": 24,
"reference": [
"121",
0
],
"reference_mix": 1.0,
"audio_source": "use pass 2's foley - ONLY IF the audio rows were seeded"
},
"_meta": {
"title": "H3AudioRecover"
}
},
"214": {
"class_type": "CreateVideo",
"inputs": {
"images": [
"213",
0
],
"audio": [
"219",
0
],
"fps": 24,
"bit_depth": 8
},
"_meta": {
"title": "CreateVideo"
}
},
"215": {
"class_type": "SaveVideo",
"inputs": {
"video": [
"214",
0
],
"format": "auto",
"codec": "auto",
"filename_prefix": "video/mainodes_ref2va_recovered"
},
"_meta": {
"title": "SaveVideo"
}
},
"9101": {
"class_type": "H3AudioSmear",
"inputs": {
"audio": [
"121",
0
],
"hold_map": [
"201",
1
],
"fps": 24
}
},
"9102": {
"class_type": "VAEEncodeAudio",
"inputs": {
"audio": [
"9101",
0
],
"vae": [
"120",
0
]
}
}
}