Seedance 2.5 live-action test: 17 s school-hallway walk, a man in a red tee leads a woman, a close-up and a turn to her at a trophy case
A long school hallway with a man in a red tee walking toward the camera and a woman behind him; the camera backs away to a close-up of his face turned aside, then turns to the woman who stops at a lit trophy case and walks away down the hall.
Task: Stage a school hallway walk in which a man in a red tee leads a woman, the camera backs away to his close-up and then turns to her at a trophy case as one continuous live-action take of 17 s.
Outcome: succeeded
0.71 s: Long school hallway: a man in a red tee walks toward the camera down the centre.
2.13 s: He walks closer; a woman in a dark top and black trousers walks behind him at the right.
3.55 s: Close: his face, looking back, mouth open.
4.97 s: Bigger close: his face turned down to the left, a trophy case behind, the woman beyond.
6.39 s: Same; his face is lowered with eyes closed.
7.81 s: Same; he looks aside.
9.23 s: Same.
10.65 s: His head turns to the right as she walks toward him.
12.07 s: He is at the left edge; she stands in the middle of the hall.
13.49 s: She stands and looks along the hall.
14.91 s: She walks away along the hallway.
16.33 s: She is far away, at the left of the trophy case.
Intent vs result: The sampled frames follow the plan: a walk, a backing close-up of the man, a trophy-case junction and the woman as the closing subject. The close-up is held for about six samples (about 10 s). The linked inputs are recorded by count only; their disclosure and rights basis are unconfirmed, and the prompt asks that they set the faces and wardrobe of the people shown, so likeness is a hold. Spoken lines were requested and not heard.
Publication update (2026-10-29): the founder has now cleared public use of this output, including incorporated input likenesses and soundtrack. Earlier rights/likeness holds described above are resolved by that confirmation. Reference-input files remain private. Audio remains not listened to; visual observations cover sampled frames only.
Structured fields
{
"inputs": [
{
"description": "9 linked image input(s) recorded in the generation metadata (details withheld)."
},
{
"description": "1 linked video input(s) recorded in the generation metadata (details withheld)."
}
],
"run_at": "2026-09-25T02:40:26Z",
"metrics": [
{
"name": "output_duration",
"unit": "s",
"value": 17.042
},
{
"name": "resolution",
"value": "1280x720"
},
{
"name": "frame_rate",
"value": "24/1"
},
{
"name": "scene_detector_transitions_thr0.30",
"value": 0
},
{
"name": "black_intervals",
"value": "none"
}
],
"procedure": "Seedance 2.5, linked reference inputs recorded in the job metadata (images that set the place, light, wardrobe and the people and one blocking video), a long structured prompt in a shot-by-shot format for one continuous live-action take with spoken lines requested, 17 s, 16:9, 720p label, audio requested, no seed set. One of a set of live-action school-scene runs on 25 September 2026 (UTC). Sampled inspection: 12 timestamped frames from 0.71 s to 16.33 s of a 17.04 s file, plus ffmpeg scene detection at threshold 0.30 and blackdetect.",
"parameters": {
"seed": "unset",
"model": "seedance_2_5 (label in job metadata)",
"multi_shots": false,
"aspect_ratio": "16:9",
"audio_requested": true,
"resolution_label": "720p",
"requested_duration_s": 17
},
"limitations": [
"Sampled visual inspection from timestamped frames (contact sheets), not full real-time playback; motion smoothness, micro-timing and lip-sync are unrated.",
"Audio was not listened to; the file has an audio track but sound, dialogue and music are unassessed.",
"Prompt intent is taken from private job metadata and summarised here only in general terms; the founder's taste and intent for each test are unknown.",
"Single historical generation without a fixed seed; not a controlled benchmark and not validation of any current product or pipeline."
],
"artifact_ids": [
"ast_01M64XW9J68NBEHFSHPCYJ3BZP"
]
}Evidence claims
- [sourced_fact] Timestamped observations above come from viewing 12 sampled frames of the output (sampled visual inspection, not full playback).
- [upstream_claim] The generation was run by the founder in an earlier test period; H01 did not run or reproduce it.
- [inference] A backing camera that lets a figure grow from medium to close-up is a clean way to isolate a face.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | unknown | permitted | permitted | unknown | unknown | Original analysis written by H01 (an AI agent) for Veyrul from private viewing; no third-party text, frames or audio reproduced. |
| media | unknown | permitted | permitted | unknown | unknown | Founder confirmed public use of all these Higgsfield outputs on 2026-10-29, including incorporated reference inputs, likenesses and soundtrack. Audio has not been listened to; permission does not establish listening or independent review. |