{
  "schema_version": 1,
  "settings": {
    "hardware": "NVIDIA GB200",
    "gpu_count": 8,
    "nodes": 2,
    "gpus_per_node": 4,
    "hardware_proxy": "B200 proxy; B200 was not separately measured",
    "canvas": "1344x768",
    "fps": 24,
    "target_seconds": [
      5,
      10,
      15
    ],
    "measured_seeds": [
      0,
      1,
      2
    ],
    "warmup_seed": 42,
    "warmup_requests_per_profile": 1,
    "concurrency": 1,
    "gallery_seed": 0,
    "latency": "client end-to-end to completed MP4;startup excluded",
    "aggregation": "arithmetic mean and sample standard deviation across three measured seeds",
    "speedup": "Base mean seconds / FastH3 Omni mean seconds; no normalization",
    "base_denoiser_evaluations": 50,
    "fast_denoiser_evaluations": 8,
    "reference_sizing": "enlarged reference image short edge 2048; native released preprocessing",
    "media": "H.264 video and AAC stereo 32kHz audio",
    "base": {
      "request_quality": "lossless",
      "reference_resize_mode": "diffusers"
    },
    "comparison_scope": "model-and-serving-system comparison; not an isolated distillation or kernel ablation",
    "quality_scope": "matching inputs and seeds do not imply numerical or perceptual output parity"
  },
  "profiles": [
    {
      "id": "image-5s",
      "composition": "image",
      "label": "1 image",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n\nsummary:\n[reference generation] <Subject 1> makes breakfast in a sunlit kitchen and speaks warmly to the camera.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm morning light.\n[Shot 1] A medium close-up with a slow push-in. In a sunlit kitchen, <Subject 1> (S1) sets a bowl of oranges on the wooden counter, wipes his hands on a striped towel, looks into the camera and says with a relaxed smile, <d>[English] Breakfast is on me today, so take your time.</d> Morning light falls across his shoulders and the steam of a kettle drifts behind him.\n\noverall_soundscape:\nBright kitchen room tone, a kettle beginning to whistle, and birdsong drifting through an open window.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-5s-seed0.mp4",
          "poster": "media/posters/base-image-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-5s-seed0.mp4",
          "poster": "media/posters/fast-image-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-5s-seed0.mp4",
          "poster": "media/posters/decoder36-image-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 35.43855305567073,
          "sample_std_s": 0.058240637634473194,
          "measured_seconds": [
            35.37135303701507,
            35.46989696403034,
            35.47440916596679
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 5.66829071372437,
          "sample_std_s": 0.01889530389227691,
          "measured_seconds": [
            5.655375283909962,
            5.659519303124398,
            5.68997755413875
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.2520704821764
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "image-10s",
      "composition": "image",
      "label": "1 image",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n\nsummary:\n[reference generation] <Subject 1> makes breakfast in a sunlit kitchen and speaks warmly to the camera.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm morning light.\n[Shot 1] A medium close-up with a slow push-in. In a sunlit kitchen, <Subject 1> (S1) sets a bowl of oranges on the wooden counter, wipes his hands on a striped towel, looks into the camera and says with a relaxed smile, <d>[English] Breakfast is on me today, so take your time.</d> Morning light falls across his shoulders and the steam of a kettle drifts behind him.\n[Shot 2] At 00:05.000, the shot cuts to a close shot over the counter. <Subject 1> (S1) slices an orange in half on a wooden board, squeezes it into a glass and slides the glass toward the camera, saying, <d>[English] Fresh juice first.</d> A drop runs down the glass in the sunlight.\n\noverall_soundscape:\nBright kitchen room tone, a kettle beginning to whistle, and birdsong drifting through an open window.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-10s-seed0.mp4",
          "poster": "media/posters/base-image-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-10s-seed0.mp4",
          "poster": "media/posters/fast-image-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-10s-seed0.mp4",
          "poster": "media/posters/decoder36-image-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 78.53238428733312,
          "sample_std_s": 0.013348087854205003,
          "measured_seconds": [
            78.5455206540064,
            78.53279810602544,
            78.51883410196751
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 9.37390313623473,
          "sample_std_s": 0.012632052208678385,
          "measured_seconds": [
            9.385046843905002,
            9.360180694842711,
            9.376481869956478
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 8.377767846113851
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "image-15s",
      "composition": "image",
      "label": "1 image",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n\nsummary:\n[reference generation] <Subject 1> makes breakfast in a sunlit kitchen and speaks warmly to the camera.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm morning light.\n[Shot 1] A medium close-up with a slow push-in. In a sunlit kitchen, <Subject 1> (S1) sets a bowl of oranges on the wooden counter, wipes his hands on a striped towel, looks into the camera and says with a relaxed smile, <d>[English] Breakfast is on me today, so take your time.</d> Morning light falls across his shoulders and the steam of a kettle drifts behind him.\n[Shot 2] At 00:05.000, the shot cuts to a close shot over the counter. <Subject 1> (S1) slices an orange in half on a wooden board, squeezes it into a glass and slides the glass toward the camera, saying, <d>[English] Fresh juice first.</d> A drop runs down the glass in the sunlight.\n[Shot 3] At 00:10.000, the shot cuts to a medium wide shot of the whole kitchen. <Subject 1> lifts the whistling kettle off the stove, pours hot water into a teapot and turns back to the camera with a grin as the steam rises past the open window.\n\noverall_soundscape:\nBright kitchen room tone, a kettle beginning to whistle, and birdsong drifting through an open window.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-15s-seed0.mp4",
          "poster": "media/posters/base-image-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-15s-seed0.mp4",
          "poster": "media/posters/fast-image-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-15s-seed0.mp4",
          "poster": "media/posters/decoder36-image-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 139.69463913034028,
          "sample_std_s": 0.11786345437191431,
          "measured_seconds": [
            139.73413636902114,
            139.78768133500125,
            139.5620996869984
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 14.141940217077112,
          "sample_std_s": 0.008489523078824073,
          "measured_seconds": [
            14.14658638718538,
            14.147092547034845,
            14.132141717011109
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 9.878039150642985
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "image-voice-5s",
      "composition": "image-voice",
      "label": "Image + voice",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Audio 1> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> finishes cooking dinner in a kitchen and speaks to the camera in the voice timbre of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Audio 1>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic live-action style with soft evening light.\n[Shot 1] A medium close-up with a slow push-in. In a warm kitchen, <Subject 1> (S1) dries his hands on a towel, looks into the camera and says, using the voice timbre referenced from <Audio 1>, <d>[English] Dinner is almost ready, come sit down.</d>\n\noverall_soundscape:\nQuiet kitchen room tone with a faint refrigerator hum, a simmering pot and the clink of plates.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-voice-5s-seed0.mp4",
          "poster": "media/posters/base-image-voice-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-voice-5s-seed0.mp4",
          "poster": "media/posters/fast-image-voice-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-voice-5s-seed0.mp4",
          "poster": "media/posters/decoder36-image-voice-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 35.922478002340846,
          "sample_std_s": 0.0658791712909475,
          "measured_seconds": [
            35.88077245501336,
            35.88823494402459,
            35.998426607984584
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 5.84164546111909,
          "sample_std_s": 0.0092929772293238,
          "measured_seconds": [
            5.840988451149315,
            5.832698424113914,
            5.8512495080940425
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.149376616816991
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "image-voice-10s",
      "composition": "image-voice",
      "label": "Image + voice",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Audio 1> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> finishes cooking dinner in a kitchen and speaks to the camera in the voice timbre of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Audio 1>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic live-action style with soft evening light.\n[Shot 1] A medium close-up with a slow push-in. In a warm kitchen, <Subject 1> (S1) dries his hands on a towel, looks into the camera and says, using the voice timbre referenced from <Audio 1>, <d>[English] Dinner is almost ready, come sit down.</d>\n[Shot 2] At 00:05.000, the shot cuts to a close shot of the stove. <Subject 1> (S1) lifts the lid of a pot, tastes the sauce from a wooden spoon, nods and says in the same voice, <d>[English] A little more salt, and it is perfect.</d> Steam curls up past his face.\n\noverall_soundscape:\nQuiet kitchen room tone with a faint refrigerator hum, a simmering pot and the clink of plates.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-voice-10s-seed0.mp4",
          "poster": "media/posters/base-image-voice-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-voice-10s-seed0.mp4",
          "poster": "media/posters/fast-image-voice-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-voice-10s-seed0.mp4",
          "poster": "media/posters/decoder36-image-voice-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 79.92020194465294,
          "sample_std_s": 0.058157222712808376,
          "measured_seconds": [
            79.8530666679726,
            79.95238990098005,
            79.95514926500618
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 9.712229180304954,
          "sample_std_s": 0.14528203992712482,
          "measured_seconds": [
            9.632657385896891,
            9.879913937067613,
            9.624116217950359
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 8.228821670180515
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "image-voice-15s",
      "composition": "image-voice",
      "label": "Image + voice",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Audio 1> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> finishes cooking dinner in a kitchen and speaks to the camera in the voice timbre of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Audio 1>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic live-action style with soft evening light.\n[Shot 1] A medium close-up with a slow push-in. In a warm kitchen, <Subject 1> (S1) dries his hands on a towel, looks into the camera and says, using the voice timbre referenced from <Audio 1>, <d>[English] Dinner is almost ready, come sit down.</d>\n[Shot 2] At 00:05.000, the shot cuts to a close shot of the stove. <Subject 1> (S1) lifts the lid of a pot, tastes the sauce from a wooden spoon, nods and says in the same voice, <d>[English] A little more salt, and it is perfect.</d> Steam curls up past his face.\n[Shot 3] At 00:10.000, the shot cuts to a medium wide shot. <Subject 1> (S1) carries two plates to a small table by the window, sets them down, looks back at the camera and says, <d>[English] Come on, before it gets cold.</d>\n\noverall_soundscape:\nQuiet kitchen room tone with a faint refrigerator hum, a simmering pot and the clink of plates.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-image-voice-15s-seed0.mp4",
          "poster": "media/posters/base-image-voice-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-image-voice-15s-seed0.mp4",
          "poster": "media/posters/fast-image-voice-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-image-voice-15s-seed0.mp4",
          "poster": "media/posters/decoder36-image-voice-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 140.91230595269008,
          "sample_std_s": 0.06212420050404835,
          "measured_seconds": [
            140.98403667705134,
            140.87577450100798,
            140.8771066800109
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 14.435537573959058,
          "sample_std_s": 0.0025606747585788492,
          "measured_seconds": [
            14.434969826834276,
            14.43833447410725,
            14.433308420935646
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 9.761486555712922
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "images3-5s",
      "composition": "images3",
      "label": "3 images",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the pair of silver wraparound sunglasses shown from three angles in <Picture 2>.\n<Subject 3> is the environment in <Picture 3>: a warmly lit traditional Japanese dining room with a family gathered around a low table.\n\nsummary:\n[reference generation] <Subject 1>, wearing <Subject 2>, joins the family dinner in <Subject 3> and shares the meal.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the silver wraparound frame and dark lenses of the sunglasses are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - the paper doors, low wooden table, warm lantern light and the family are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm lantern light.\n[Shot 1] A medium wide shot with a gentle handheld drift. <Subject 1> (S1) slides open the paper door of <Subject 3>, with <Subject 2> pushed up on his forehead. He smiles, lowers the sunglasses, bows slightly and says, <d>[English] Sorry I'm late, the train was packed.</d> The family laughs and makes room for him at the low table while steam rises from the bowls of ramen.\n\noverall_soundscape:\nSoft family chatter, clinking ceramic bowls, the slide of a paper door and the hiss of a simmering pot.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/hf_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/official_fl2va_0_image.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images3-5s-seed0.mp4",
          "poster": "media/posters/base-images3-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images3-5s-seed0.mp4",
          "poster": "media/posters/fast-images3-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images3-5s-seed0.mp4",
          "poster": "media/posters/decoder36-images3-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 73.09455884131603,
          "sample_std_s": 0.9769753039105809,
          "measured_seconds": [
            72.34318519500084,
            72.7415077299811,
            74.19898359896615
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 12.421142409710834,
          "sample_std_s": 0.006399473759445845,
          "measured_seconds": [
            12.416745302034542,
            12.418197778984904,
            12.428484148113057
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 5.884688898194323
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "images3-10s",
      "composition": "images3",
      "label": "3 images",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the pair of silver wraparound sunglasses shown from three angles in <Picture 2>.\n<Subject 3> is the environment in <Picture 3>: a warmly lit traditional Japanese dining room with a family gathered around a low table.\n\nsummary:\n[reference generation] <Subject 1>, wearing <Subject 2>, joins the family dinner in <Subject 3> and shares the meal.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - the silver wraparound frame and dark lenses of the sunglasses are retained.\n<Subject 3> (appears in [Shot 1], [Shot 2]): fully_preserved - the paper doors, low wooden table, warm lantern light and the family are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm lantern light.\n[Shot 1] A medium wide shot with a gentle handheld drift. <Subject 1> (S1) slides open the paper door of <Subject 3>, with <Subject 2> pushed up on his forehead. He smiles, lowers the sunglasses, bows slightly and says, <d>[English] Sorry I'm late, the train was packed.</d> The family laughs and makes room for him at the low table while steam rises from the bowls of ramen.\n[Shot 2] At 00:05.000, the shot cuts to a medium shot across the table. <Subject 1> kneels at the low table, takes off <Subject 2> and sets them beside his bowl. The child across from him picks up the sunglasses, tries them on, and the whole family laughs.\n\noverall_soundscape:\nSoft family chatter, clinking ceramic bowls, the slide of a paper door and the hiss of a simmering pot.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/hf_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/official_fl2va_0_image.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images3-10s-seed0.mp4",
          "poster": "media/posters/base-images3-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images3-10s-seed0.mp4",
          "poster": "media/posters/fast-images3-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images3-10s-seed0.mp4",
          "poster": "media/posters/decoder36-images3-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 131.25289688501894,
          "sample_std_s": 0.01982337014850291,
          "measured_seconds": [
            131.24267110199435,
            131.2402744720457,
            131.2757450810168
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 19.152300127083436,
          "sample_std_s": 0.05856982268760325,
          "measured_seconds": [
            19.122103007975966,
            19.219805982196704,
            19.114991391077638
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.853114039259079
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "images3-15s",
      "composition": "images3",
      "label": "3 images",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the pair of silver wraparound sunglasses shown from three angles in <Picture 2>.\n<Subject 3> is the environment in <Picture 3>: a warmly lit traditional Japanese dining room with a family gathered around a low table.\n\nsummary:\n[reference generation] <Subject 1>, wearing <Subject 2>, joins the family dinner in <Subject 3> and shares the meal.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - the silver wraparound frame and dark lenses of the sunglasses are retained.\n<Subject 3> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the paper doors, low wooden table, warm lantern light and the family are retained.\n\ndetailed_description:\nThe target video is in a realistic live-action style with warm lantern light.\n[Shot 1] A medium wide shot with a gentle handheld drift. <Subject 1> (S1) slides open the paper door of <Subject 3>, with <Subject 2> pushed up on his forehead. He smiles, lowers the sunglasses, bows slightly and says, <d>[English] Sorry I'm late, the train was packed.</d> The family laughs and makes room for him at the low table while steam rises from the bowls of ramen.\n[Shot 2] At 00:05.000, the shot cuts to a medium shot across the table. <Subject 1> kneels at the low table, takes off <Subject 2> and sets them beside his bowl. The child across from him picks up the sunglasses, tries them on, and the whole family laughs.\n[Shot 3] At 00:10.000, the shot cuts to a close-up of a steaming bowl of ramen. <Subject 1> (S1) lifts noodles with his chopsticks, takes a bite and says, <d>[English] This is worth the trip.</d> The lantern light glows on the lacquered table.\n\noverall_soundscape:\nSoft family chatter, clinking ceramic bowls, the slide of a paper door and the hiss of a simmering pot.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/hf_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/official_fl2va_0_image.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images3-15s-seed0.mp4",
          "poster": "media/posters/base-images3-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images3-15s-seed0.mp4",
          "poster": "media/posters/fast-images3-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images3-15s-seed0.mp4",
          "poster": "media/posters/decoder36-images3-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 206.8765430636704,
          "sample_std_s": 0.138467136439225,
          "measured_seconds": [
            206.7778816630016,
            206.81691233097808,
            207.0348351970315
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 26.532887148981292,
          "sample_std_s": 0.01347773797958985,
          "measured_seconds": [
            26.517897497862577,
            26.536758138099685,
            26.544005810981616
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 7.796985752137164
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "images9-5s",
      "composition": "images9",
      "label": "9 images",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n\nsummary:\n[reference generation] A five-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 4]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 5]): fully_preserved - the great window and the starfield of the observation deck are retained.\n<Subject 8> (appears in [Shot 5]): weak_reference - the closing card follows its flat noir palette and lettering.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:01.000, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:02.000, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:03.000, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light.\n[Shot 5] At 00:04.000, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images9-5s-seed0.mp4",
          "poster": "media/posters/base-images9-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images9-5s-seed0.mp4",
          "poster": "media/posters/fast-images9-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images9-5s-seed0.mp4",
          "poster": "media/posters/decoder36-images9-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 121.61350872767314,
          "sample_std_s": 0.046004131898873424,
          "measured_seconds": [
            121.59475235798163,
            121.57984589604894,
            121.66592792898882
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 21.588712421633925,
          "sample_std_s": 0.01170487817517591,
          "measured_seconds": [
            21.575838191900402,
            21.59871251694858,
            21.59158655605279
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 5.633198791689167
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "images9-10s",
      "composition": "images9",
      "label": "9 images",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n\nsummary:\n[reference generation] A five-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 4]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 5]): fully_preserved - the great window and the starfield of the observation deck are retained.\n<Subject 8> (appears in [Shot 5]): weak_reference - the closing card follows its flat noir palette and lettering.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:02.000, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:04.000, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:06.000, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light.\n[Shot 5] At 00:08.000, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images9-10s-seed0.mp4",
          "poster": "media/posters/base-images9-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images9-10s-seed0.mp4",
          "poster": "media/posters/fast-images9-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images9-10s-seed0.mp4",
          "poster": "media/posters/decoder36-images9-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 193.94788071833318,
          "sample_std_s": 0.05444215624088368,
          "measured_seconds": [
            193.8873188960133,
            193.96356205298798,
            193.99276120599825
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 30.795576869665336,
          "sample_std_s": 0.014092646849861525,
          "measured_seconds": [
            30.794670596020296,
            30.781959231942892,
            30.810100781032816
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.297913545804634
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "images9-15s",
      "composition": "images9",
      "label": "9 images",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n\nsummary:\n[reference generation] A five-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 4]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 5]): fully_preserved - the great window and the starfield of the observation deck are retained.\n<Subject 8> (appears in [Shot 5]): weak_reference - the closing card follows its flat noir palette and lettering.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:03.000, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:06.000, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:09.000, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light.\n[Shot 5] At 00:12.000, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-images9-15s-seed0.mp4",
          "poster": "media/posters/base-images9-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-images9-15s-seed0.mp4",
          "poster": "media/posters/fast-images9-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-images9-15s-seed0.mp4",
          "poster": "media/posters/decoder36-images9-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 282.81492259803537,
          "sample_std_s": 0.05242696851702442,
          "measured_seconds": [
            282.75438981404295,
            282.844538164034,
            282.8458398160292
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 41.56994981404083,
          "sample_std_s": 0.010942846546245187,
          "measured_seconds": [
            41.57516728108749,
            41.557374665979296,
            41.57730749505572
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.8033501089893225
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "clip-5s",
      "composition": "clip",
      "label": "1 short clip",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clip-5s-seed0.mp4",
          "poster": "media/posters/base-clip-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clip-5s-seed0.mp4",
          "poster": "media/posters/fast-clip-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clip-5s-seed0.mp4",
          "poster": "media/posters/decoder36-clip-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 36.062882283004,
          "sample_std_s": 0.12518041700765417,
          "measured_seconds": [
            35.94807549403049,
            36.044228543003555,
            36.196342811977956
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 5.1663130777888,
          "sample_std_s": 0.005713337080356295,
          "measured_seconds": [
            5.169865147210658,
            5.169351533986628,
            5.159722552169114
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.980390413822741
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "clip-10s",
      "composition": "clip",
      "label": "1 short clip",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n[Shot 2] At 00:05.000, the shot cuts to a wide shot from down the hill. <Subject 1> walks slowly through the flock with the black lamb in his arms, the white lambs parting around his legs and the long grass bending in the breeze.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clip-10s-seed0.mp4",
          "poster": "media/posters/base-clip-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clip-10s-seed0.mp4",
          "poster": "media/posters/fast-clip-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clip-10s-seed0.mp4",
          "poster": "media/posters/decoder36-clip-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 79.73047793000781,
          "sample_std_s": 0.018366907796284096,
          "measured_seconds": [
            79.72240123804659,
            79.7514991670032,
            79.71753338497365
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 8.062355255708098,
          "sample_std_s": 0.0069079275843772605,
          "measured_seconds": [
            8.061800054972991,
            8.055741682182997,
            8.069524029968306
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 9.889229065360164
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "clip-15s",
      "composition": "clip",
      "label": "1 short clip",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n[Shot 2] At 00:05.000, the shot cuts to a wide shot from down the hill. <Subject 1> walks slowly through the flock with the black lamb in his arms, the white lambs parting around his legs and the long grass bending in the breeze.\n[Shot 3] At 00:10.000, the shot cuts to a close shot near the ground. <Subject 1> kneels and sets the black lamb down among the white lambs; it shakes its ears and trots off, and he watches it go with a smile, his rings catching the light.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clip-15s-seed0.mp4",
          "poster": "media/posters/base-clip-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clip-15s-seed0.mp4",
          "poster": "media/posters/fast-clip-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clip-15s-seed0.mp4",
          "poster": "media/posters/decoder36-clip-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 140.29390113899717,
          "sample_std_s": 0.16635365778268807,
          "measured_seconds": [
            140.19495453697164,
            140.4859602630022,
            140.20078861701768
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 11.762368205313882,
          "sample_std_s": 0.010254680097150063,
          "measured_seconds": [
            11.75392372906208,
            11.759401801042259,
            11.773779085837305
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 11.927351591970794
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "video-5s",
      "composition": "video",
      "label": "1 long clip",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-video-5s-seed0.mp4",
          "poster": "media/posters/base-video-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-video-5s-seed0.mp4",
          "poster": "media/posters/fast-video-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-video-5s-seed0.mp4",
          "poster": "media/posters/decoder36-video-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 69.92695426667342,
          "sample_std_s": 0.002800567660624819,
          "measured_seconds": [
            69.92412307101768,
            69.92701656202553,
            69.92972316697706
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 8.336950273020193,
          "sample_std_s": 0.004138709271642053,
          "measured_seconds": [
            8.341374794952571,
            8.336302179144695,
            8.333173844963312
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 8.38759402139762
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "video-10s",
      "composition": "video",
      "label": "1 long clip",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n[Shot 2] At 00:05.000, the shot cuts to a wide shot from down the hill. <Subject 1> walks slowly through the flock with the black lamb in his arms, the white lambs parting around his legs and the long grass bending in the breeze.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-video-10s-seed0.mp4",
          "poster": "media/posters/base-video-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-video-10s-seed0.mp4",
          "poster": "media/posters/fast-video-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-video-10s-seed0.mp4",
          "poster": "media/posters/decoder36-video-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 139.2705134356705,
          "sample_std_s": 0.02484170062159256,
          "measured_seconds": [
            139.29687754396582,
            139.2671196620213,
            139.24754310102435
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 13.506877736654133,
          "sample_std_s": 0.03160007395520295,
          "measured_seconds": [
            13.481284708017483,
            13.497150655835867,
            13.542197846109048
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 10.311081224769419
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "video-15s",
      "composition": "video",
      "label": "1 long clip",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n\nsummary:\n[reference generation + audio reference] <Subject 1> spends a quiet moment with the black lamb in <Subject 2>, over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the man's blonde hair, pink suit, white shirt, rings and the black lamb are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced throughout.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot. <Subject 1> cradles the black lamb in the warm, low sunlight of <Subject 2>. The lamb lifts its head and nuzzles his chin; he laughs softly, tilts his head toward the camera and grins while the white lambs keep grazing on the hill.\n[Shot 2] At 00:05.000, the shot cuts to a wide shot from down the hill. <Subject 1> walks slowly through the flock with the black lamb in his arms, the white lambs parting around his legs and the long grass bending in the breeze.\n[Shot 3] At 00:10.000, the shot cuts to a close shot near the ground. <Subject 1> kneels and sets the black lamb down among the white lambs; it shakes its ears and trots off, and he watches it go with a smile, his rings catching the light.\n\noverall_soundscape:\nA light breeze over the grass and distant bleating, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-video-15s-seed0.mp4",
          "poster": "media/posters/base-video-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-video-15s-seed0.mp4",
          "poster": "media/posters/fast-video-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-video-15s-seed0.mp4",
          "poster": "media/posters/decoder36-video-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 215.74971013966328,
          "sample_std_s": 0.16778376777258394,
          "measured_seconds": [
            215.6278522890061,
            215.68019984196872,
            215.94107828801498
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 18.407617761287838,
          "sample_std_s": 0.04033983759025736,
          "measured_seconds": [
            18.453869292978197,
            18.389277322916314,
            18.379706667969003
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 11.72067526268369
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "clips3-5s",
      "composition": "clips3",
      "label": "3 clips",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the white robotic arm with a black gripper in <Video 2>.\n<Subject 3> is the dark starship observation deck in <Video 3>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A three-shot sequence shows <Subject 1> in a golden pasture, <Subject 2> at work on a white table and the great window of <Subject 3>, with sound in the manner of <Audio 1>, <Audio 2> and <Audio 3>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the dark deck, its great window and the starfield are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the first shot.\n<Audio 2>: reference - the servo whir is referenced in the second shot.\n<Audio 3>: reference - the low starship hum is referenced in the third shot.\n\ndetailed_description:\nThe target video is in a realistic photographic style.\n[Shot 1] <Subject 1> holds the black lamb in a golden pasture and looks up at the sky while soft music in the manner of <Audio 1> plays.\n[Shot 2] At 00:01.667, the shot cuts to a white table, where <Subject 2> grips a red cube and drops it into a glass jar with a servo whir like <Audio 2>.\n[Shot 3] At 00:03.333, the shot cuts to <Subject 3>: the camera pushes in toward the great window, where a fleet of ships drifts past the stars over a low hum like <Audio 3>.\n\noverall_soundscape:\nThe pasture's soft music gives way to a servo whir and then a low starship hum.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clips3-5s-seed0.mp4",
          "poster": "media/posters/base-clips3-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clips3-5s-seed0.mp4",
          "poster": "media/posters/fast-clips3-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clips3-5s-seed0.mp4",
          "poster": "media/posters/decoder36-clips3-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 74.34509099967545,
          "sample_std_s": 0.0861188631937047,
          "measured_seconds": [
            74.28441829001531,
            74.3071955780033,
            74.44365913100773
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 9.448077250737697,
          "sample_std_s": 0.010358993461790254,
          "measured_seconds": [
            9.43644911213778,
            9.451462629018351,
            9.45632001105696
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 7.868806427664492
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "clips3-10s",
      "composition": "clips3",
      "label": "3 clips",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the white robotic arm with a black gripper in <Video 2>.\n<Subject 3> is the dark starship observation deck in <Video 3>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A three-shot sequence shows <Subject 1> in a golden pasture, <Subject 2> at work on a white table and the great window of <Subject 3>, with sound in the manner of <Audio 1>, <Audio 2> and <Audio 3>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the dark deck, its great window and the starfield are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the first shot.\n<Audio 2>: reference - the servo whir is referenced in the second shot.\n<Audio 3>: reference - the low starship hum is referenced in the third shot.\n\ndetailed_description:\nThe target video is in a realistic photographic style.\n[Shot 1] <Subject 1> holds the black lamb in a golden pasture and looks up at the sky while soft music in the manner of <Audio 1> plays.\n[Shot 2] At 00:03.333, the shot cuts to a white table, where <Subject 2> grips a red cube and drops it into a glass jar with a servo whir like <Audio 2>.\n[Shot 3] At 00:06.667, the shot cuts to <Subject 3>: the camera pushes in toward the great window, where a fleet of ships drifts past the stars over a low hum like <Audio 3>.\n\noverall_soundscape:\nThe pasture's soft music gives way to a servo whir and then a low starship hum.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clips3-10s-seed0.mp4",
          "poster": "media/posters/base-clips3-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clips3-10s-seed0.mp4",
          "poster": "media/posters/fast-clips3-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clips3-10s-seed0.mp4",
          "poster": "media/posters/decoder36-clips3-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 132.62300849001622,
          "sample_std_s": 0.06212935872163365,
          "measured_seconds": [
            132.69381296104984,
            132.57760113698896,
            132.59761137200985
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 13.549096170347184,
          "sample_std_s": 0.03993070410644651,
          "measured_seconds": [
            13.592919077957049,
            13.514769979054108,
            13.539599454030395
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 9.788328817110896
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "clips3-15s",
      "composition": "clips3",
      "label": "3 clips",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 2> is the white robotic arm with a black gripper in <Video 2>.\n<Subject 3> is the dark starship observation deck in <Video 3>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A three-shot sequence shows <Subject 1> in a golden pasture, <Subject 2> at work on a white table and the great window of <Subject 3>, with sound in the manner of <Audio 1>, <Audio 2> and <Audio 3>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the white robotic arm and its black gripper are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the dark deck, its great window and the starfield are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the first shot.\n<Audio 2>: reference - the servo whir is referenced in the second shot.\n<Audio 3>: reference - the low starship hum is referenced in the third shot.\n\ndetailed_description:\nThe target video is in a realistic photographic style.\n[Shot 1] <Subject 1> holds the black lamb in a golden pasture and looks up at the sky while soft music in the manner of <Audio 1> plays.\n[Shot 2] At 00:05.000, the shot cuts to a white table, where <Subject 2> grips a red cube and drops it into a glass jar with a servo whir like <Audio 2>.\n[Shot 3] At 00:10.000, the shot cuts to <Subject 3>: the camera pushes in toward the great window, where a fleet of ships drifts past the stars over a low hum like <Audio 3>.\n\noverall_soundscape:\nThe pasture's soft music gives way to a servo whir and then a low starship hum.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-clips3-15s-seed0.mp4",
          "poster": "media/posters/base-clips3-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-clips3-15s-seed0.mp4",
          "poster": "media/posters/fast-clips3-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-clips3-15s-seed0.mp4",
          "poster": "media/posters/decoder36-clips3-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 206.43853406897202,
          "sample_std_s": 0.2091587000206405,
          "measured_seconds": [
            206.33237703994382,
            206.67948321899166,
            206.3037419479806
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 18.378265630997095,
          "sample_std_s": 0.005370571279043227,
          "measured_seconds": [
            18.37207352509722,
            18.381067737936974,
            18.381655629957095
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 11.232753852506587
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "mixed-5s",
      "composition": "mixed",
      "label": "Image + video + voice",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 3> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n<Audio 2> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> walks into <Subject 3> and greets <Subject 2>, speaking in the voice timbre of <Audio 2> over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 3> (appears in [Shot 1]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced beneath the dialogue.\n<Audio 2>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot of <Subject 3>. <Subject 2> cradles the black lamb in the warm sunlight as <Subject 1> (S1) steps into frame from the left, waves and says, using the voice timbre referenced from <Audio 2>, <d>[English] There you are, the whole farm is looking for that lamb.</d> <Subject 2> laughs and lifts the lamb a little higher.\n\noverall_soundscape:\nA light breeze over the grass, distant bleating and soft footsteps, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-mixed-5s-seed0.mp4",
          "poster": "media/posters/base-mixed-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-mixed-5s-seed0.mp4",
          "poster": "media/posters/fast-mixed-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-mixed-5s-seed0.mp4",
          "poster": "media/posters/decoder36-mixed-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 94.13728581630858,
          "sample_std_s": 0.07057947975904619,
          "measured_seconds": [
            94.21840859699296,
            94.08995805296581,
            94.10349079896696
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 13.060403493931517,
          "sample_std_s": 0.0032006345453233468,
          "measured_seconds": [
            13.056708194082603,
            13.062199852894992,
            13.062302434816957
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 7.207839012022043
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "mixed-10s",
      "composition": "mixed",
      "label": "Image + video + voice",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 3> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n<Audio 2> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> walks into <Subject 3> and greets <Subject 2>, speaking in the voice timbre of <Audio 2> over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 3> (appears in [Shot 1], [Shot 2]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced beneath the dialogue.\n<Audio 2>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot of <Subject 3>. <Subject 2> cradles the black lamb in the warm sunlight as <Subject 1> (S1) steps into frame from the left, waves and says, using the voice timbre referenced from <Audio 2>, <d>[English] There you are, the whole farm is looking for that lamb.</d> <Subject 2> laughs and lifts the lamb a little higher.\n[Shot 2] At 00:05.000, the shot cuts to a two-shot from the side. <Subject 1> (S1) reaches out and scratches the lamb behind its ears and says in the same voice, <d>[English] He looks far too comfortable to come home.</d> <Subject 2> grins and shakes his head while the white lambs graze behind them.\n\noverall_soundscape:\nA light breeze over the grass, distant bleating and soft footsteps, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-mixed-10s-seed0.mp4",
          "poster": "media/posters/base-mixed-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-mixed-10s-seed0.mp4",
          "poster": "media/posters/fast-mixed-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-mixed-10s-seed0.mp4",
          "poster": "media/posters/decoder36-mixed-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 172.37049457599642,
          "sample_std_s": 0.09582312516330484,
          "measured_seconds": [
            172.47736188396811,
            172.34189245803282,
            172.29222938598832
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 19.747013999071594,
          "sample_std_s": 0.0031701915640889398,
          "measured_seconds": [
            19.744587775086984,
            19.750600967090577,
            19.74585325503722
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 8.72893970623105
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "mixed-15s",
      "composition": "mixed",
      "label": "Image + video + voice",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Subject 3> is the environment in <Video 1>: a grassy hill at golden hour with white lambs grazing.\n<Audio 1> is the synchronized audio track of <Video 1>, a soft background music.\n<Audio 2> is the voice-timbre reference for <Subject 1> (S1).\n\nsummary:\n[reference generation + audio reference] <Subject 1> walks into <Subject 3> and greets <Subject 2>, speaking in the voice timbre of <Audio 2> over music in the manner of <Audio 1>.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Subject 3> (appears in [Shot 1], [Shot 2], [Shot 3]): fully_preserved - the grassy hill, golden light and grazing lambs are retained.\n<Audio 1>: reference - the soft background music is referenced beneath the dialogue.\n<Audio 2>: reference - its vocal timbre guides the dialogue delivery of <Subject 1> without copying the original signal.\n\ndetailed_description:\nThe target video is in a realistic photographic style with golden-hour light.\n[Shot 1] A steady medium shot of <Subject 3>. <Subject 2> cradles the black lamb in the warm sunlight as <Subject 1> (S1) steps into frame from the left, waves and says, using the voice timbre referenced from <Audio 2>, <d>[English] There you are, the whole farm is looking for that lamb.</d> <Subject 2> laughs and lifts the lamb a little higher.\n[Shot 2] At 00:05.000, the shot cuts to a two-shot from the side. <Subject 1> (S1) reaches out and scratches the lamb behind its ears and says in the same voice, <d>[English] He looks far too comfortable to come home.</d> <Subject 2> grins and shakes his head while the white lambs graze behind them.\n[Shot 3] At 00:10.000, the shot cuts to a wide shot from down the hill. <Subject 1> and <Subject 2> walk side by side toward the camera through the flock, the black lamb still in <Subject 2>'s arms, their long shadows stretching across the grass.\n\noverall_soundscape:\nA light breeze over the grass, distant bleating and soft footsteps, beneath gentle music in the manner of <Audio 1>.\n\nnon_diegetic_music:\nN/A",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/hf_character_action_reference.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/official_ref2va_0_video.mp4"
        },
        {
          "type": "audio",
          "caption": "Audio 1",
          "file": "media/inputs/official_ref2va_1_audio.mp3"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-mixed-15s-seed0.mp4",
          "poster": "media/posters/base-mixed-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-mixed-15s-seed0.mp4",
          "poster": "media/posters/fast-mixed-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-mixed-15s-seed0.mp4",
          "poster": "media/posters/decoder36-mixed-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 257.70798912632745,
          "sample_std_s": 0.06287601855036154,
          "measured_seconds": [
            257.7749586729915,
            257.69878816598793,
            257.6502205400029
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 26.71353052626364,
          "sample_std_s": 0.005738865692638722,
          "measured_seconds": [
            26.718373663024977,
            26.715025920886546,
            26.707191994879395
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 9.64709583680673
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    },
    {
      "id": "full-5s",
      "composition": "full",
      "label": "9 images + 3 clips",
      "target_seconds": 5,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9> and <Video 2>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7> and <Video 3>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n<Subject 9> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A six-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a golden pasture, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5], [Shot 6]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 5]): fully_preserved - the white robotic arm and its motion are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 6]): fully_preserved - the great window and the drifting fleet are retained.\n<Subject 8> (appears in [Shot 6]): weak_reference - the closing card follows its flat noir palette and lettering.\n<Subject 9> (appears in [Shot 4]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the fourth shot.\n<Audio 2>: reference - the servo whir is referenced in the fifth shot.\n<Audio 3>: reference - the low starship hum is referenced in the sixth shot.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:00.833, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:01.667, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:02.500, the shot cuts to a golden pasture, where <Subject 9> slides <Subject 4> onto his nose with one hand while holding the black lamb, over soft music in the manner of <Audio 1>.\n[Shot 5] At 00:03.333, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light with a servo whir like <Audio 2>.\n[Shot 6] At 00:04.167, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4> before the great window over a low hum like <Audio 3>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a breeze over grass, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-full-5s-seed0.mp4",
          "poster": "media/posters/base-full-5s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-full-5s-seed0.mp4",
          "poster": "media/posters/fast-full-5s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-full-5s-seed0.mp4",
          "poster": "media/posters/decoder36-full-5s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 219.89730588699845,
          "sample_std_s": 0.13630394775413066,
          "measured_seconds": [
            219.74091394298011,
            219.96017423499143,
            219.99082948302384
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 37.38071476706924,
          "sample_std_s": 0.03122306025679131,
          "measured_seconds": [
            37.348596648080274,
            37.41095830104314,
            37.382589352084324
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 5.882640480719707
      },
      "frames": 124,
      "actual_video_seconds": 5.166666666666667
    },
    {
      "id": "full-10s",
      "composition": "full",
      "label": "9 images + 3 clips",
      "target_seconds": 10,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9> and <Video 2>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7> and <Video 3>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n<Subject 9> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A six-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a golden pasture, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5], [Shot 6]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 5]): fully_preserved - the white robotic arm and its motion are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 6]): fully_preserved - the great window and the drifting fleet are retained.\n<Subject 8> (appears in [Shot 6]): weak_reference - the closing card follows its flat noir palette and lettering.\n<Subject 9> (appears in [Shot 4]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the fourth shot.\n<Audio 2>: reference - the servo whir is referenced in the fifth shot.\n<Audio 3>: reference - the low starship hum is referenced in the sixth shot.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:01.667, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:03.333, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:05.000, the shot cuts to a golden pasture, where <Subject 9> slides <Subject 4> onto his nose with one hand while holding the black lamb, over soft music in the manner of <Audio 1>.\n[Shot 5] At 00:06.667, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light with a servo whir like <Audio 2>.\n[Shot 6] At 00:08.333, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4> before the great window over a low hum like <Audio 3>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a breeze over grass, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-full-10s-seed0.mp4",
          "poster": "media/posters/base-full-10s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-full-10s-seed0.mp4",
          "poster": "media/posters/fast-full-10s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-full-10s-seed0.mp4",
          "poster": "media/posters/decoder36-full-10s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 314.15691652236274,
          "sample_std_s": 0.13206951664013808,
          "measured_seconds": [
            314.29475149605423,
            314.0314879990183,
            314.1445100720157
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 48.37904121199002,
          "sample_std_s": 0.03887000946685597,
          "measured_seconds": [
            48.35105221392587,
            48.42342221504077,
            48.362649207003415
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 6.493657349383428
      },
      "frames": 243,
      "actual_video_seconds": 10.125
    },
    {
      "id": "full-15s",
      "composition": "full",
      "label": "9 images + 3 clips",
      "target_seconds": 15,
      "seed": 0,
      "width": 1344,
      "height": 768,
      "fps": 24,
      "gpu_count": 8,
      "prompt": "subject_definitions:\n<Subject 1> is the young man with dark curly hair wearing a white T-shirt in <Picture 1>.\n<Subject 2> is the model in the black sleeveless dress in <Picture 3> and <Picture 4>.\n<Subject 3> is the clay fox in <Picture 6>.\n<Subject 4> is the pair of silver wraparound sunglasses in <Picture 5>.\n<Subject 5> is the white robotic arm in <Picture 9> and <Video 2>.\n<Subject 6> is the family dining room in <Picture 2>.\n<Subject 7> is the starship observation deck in <Picture 7> and <Video 3>.\n<Subject 8> is the noir title-card style of <Picture 8>.\n<Subject 9> is the young man with short wavy blonde hair, wearing a bright pink suit jacket, matching pink trousers, an unbuttoned white shirt and silver rings, holding a small black lamb in his arms in <Video 1>.\n<Audio 1>, <Audio 2> and <Audio 3> are the synchronized audio tracks of <Video 1>, <Video 2> and <Video 3>.\n\nsummary:\n[reference generation + audio reference] A six-shot sunglasses commercial cuts between a family dinner, a fashion studio, a clay forest, a golden pasture, a robot lab and a starship deck, and ends on a noir title card.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - the young man's face, dark curly hair and white T-shirt are retained.\n<Subject 2> (appears in [Shot 2]): fully_preserved - the model's braided hair and black sleeveless dress are retained.\n<Subject 3> (appears in [Shot 3]): fully_preserved - the clay fox's shape and stop-motion texture are retained.\n<Subject 4> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5], [Shot 6]): fully_preserved - the silver wraparound frame and dark lenses are retained.\n<Subject 5> (appears in [Shot 5]): fully_preserved - the white robotic arm and its motion are retained.\n<Subject 6> (appears in [Shot 1]): fully_preserved - the low table and warm light of the dining room are retained.\n<Subject 7> (appears in [Shot 6]): fully_preserved - the great window and the drifting fleet are retained.\n<Subject 8> (appears in [Shot 6]): weak_reference - the closing card follows its flat noir palette and lettering.\n<Subject 9> (appears in [Shot 4]): fully_preserved - the man's blonde hair, pink suit and the black lamb are retained.\n<Audio 1>: reference - the pasture's soft music is referenced in the fourth shot.\n<Audio 2>: reference - the servo whir is referenced in the fifth shot.\n<Audio 3>: reference - the low starship hum is referenced in the sixth shot.\n\ndetailed_description:\nThe target video is a fast-cut commercial that changes visual style with each shot.\n[Shot 1] A live-action medium shot. <Subject 1> sits down at the table of <Subject 6> and puts on <Subject 4>.\n[Shot 2] At 00:02.500, the shot cuts to a white-cyclorama fashion studio, where <Subject 2> turns toward the camera wearing <Subject 4>.\n[Shot 3] At 00:05.000, the shot cuts to stop-motion clay animation: <Subject 3> finds <Subject 4> on a mossy stump and tries them on.\n[Shot 4] At 00:07.500, the shot cuts to a golden pasture, where <Subject 9> slides <Subject 4> onto his nose with one hand while holding the black lamb, over soft music in the manner of <Audio 1>.\n[Shot 5] At 00:10.000, the shot cuts to a white table, where <Subject 5> lifts <Subject 4> and holds them up to the light with a servo whir like <Audio 2>.\n[Shot 6] At 00:12.500, the shot cuts to <Subject 7>: a silhouetted captain slips on <Subject 4> before the great window over a low hum like <Audio 3>, and the frame cuts to a title card in the style of <Subject 8>.\n\noverall_soundscape:\nQuick cuts carry a dinner murmur, a camera shutter, rustling leaves, a breeze over grass, a servo whir and a low starship hum.\n\nnon_diegetic_music:\nA cool jazz bass line with brushed drums at a steady medium tempo.",
      "conditions": [
        {
          "type": "image",
          "caption": "Picture 1",
          "file": "media/inputs/square_character_action_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 2",
          "file": "media/inputs/square_official_fl2va_0_image.png"
        },
        {
          "type": "image",
          "caption": "Picture 3",
          "file": "media/inputs/square_fashion_glasses_reference_1.png"
        },
        {
          "type": "image",
          "caption": "Picture 4",
          "file": "media/inputs/square_fashion_glasses_reference_2.png"
        },
        {
          "type": "image",
          "caption": "Picture 5",
          "file": "media/inputs/square_fashion_glasses_reference_3.png"
        },
        {
          "type": "image",
          "caption": "Picture 6",
          "file": "media/inputs/square_fl2va_clay_fox_reference.png"
        },
        {
          "type": "image",
          "caption": "Picture 7",
          "file": "media/inputs/square_h3_cinematic_shot_3s.png"
        },
        {
          "type": "image",
          "caption": "Picture 8",
          "file": "media/inputs/square_h3_suspense_title_2s.png"
        },
        {
          "type": "image",
          "caption": "Picture 9",
          "file": "media/inputs/square_robot_arm_red_cube_1s.png"
        },
        {
          "type": "video",
          "caption": "Video 1 · with original audio",
          "file": "media/inputs/clip_official_ref2va_0_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 2 · with original audio",
          "file": "media/inputs/clip_robot_arm_red_cube_2s.mp4"
        },
        {
          "type": "video",
          "caption": "Video 3 · with original audio",
          "file": "media/inputs/clip_h3_cinematic_shot_2s.mp4"
        }
      ],
      "outputs": {
        "base": {
          "label": "Base",
          "file": "media/outputs/base-full-15s-seed0.mp4",
          "poster": "media/posters/base-full-15s-seed0.jpg"
        },
        "fast": {
          "label": "FastH3 Omni",
          "file": "media/outputs/fast-full-15s-seed0.mp4",
          "poster": "media/posters/fast-full-15s-seed0.jpg"
        },
        "decoder36": {
          "label": "FastH3 Omni · 36L VAE",
          "file": "media/outputs/decoder36-full-15s-seed0.mp4",
          "poster": "media/posters/decoder36-full-15s-seed0.jpg"
        }
      },
      "performance": {
        "base": {
          "mean_s": 422.1996387276643,
          "sample_std_s": 0.2164798269318032,
          "measured_seconds": [
            422.3884274539887,
            421.96335487201577,
            422.2471338569885
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "fast": {
          "mean_s": 59.77270269401682,
          "sample_std_s": 0.04741506772720853,
          "measured_seconds": [
            59.719835055060685,
            59.786809785990044,
            59.81146324099973
          ],
          "measured_seeds": [
            0,
            1,
            2
          ]
        },
        "speedup": 7.0634189136293815
      },
      "frames": 362,
      "actual_video_seconds": 15.083333333333334
    }
  ]
}