Seedance 2.5 sequence run with one linked image: 20 s cobalt terrace in snow, a clinch at a rail, a horned portrait and a long-haired portrait
A cobalt terrace with a lantern, a rail and falling snow: two figures clinched at the rail in a scatter of sparks, a close of a blue-skinned horned man with scaled shoulder and bared teeth, a close of a long-haired woman with golden eyes and a hand on a hilt, and a hooded figure and a horned man at the rail.
Task: Render two figures on a snowy night terrace, one horned and scaled as a cobalt night painting look, at 20 s.
Outcome: succeeded
0.84 s: A cobalt terrace in snow with a lantern: two figures clinch at a rail, a scatter of sparks.
2.51 s: Same.
4.18 s: Same.
5.85 s: Close: a blue-skinned horned man with a scaled shoulder, teeth bared.
7.52 s: Same, calmer.
9.19 s: Close: a long-haired woman with golden eyes, a hand gripping a hilt at the right.
10.86 s: Same.
12.53 s: Same.
14.20 s: Same.
15.87 s: Wide: a hooded figure and a horned man at the rail.
17.54 s: Same.
19.21 s: Same.
Intent vs result: The sampled frames follow a wide to about 5 s (shorter than the 8 s the prompt asks for), a horned man's close-up, a woman's close-up of about 6 s and a closing wide at about 15.9 to 19.2 s. One linked image input is recorded; its disclosure and rights basis are unconfirmed. At this size the palette and medium read as consistent with the linked reference image (compared locally by H01; the image is not reproduced).
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": "1 linked image input(s) recorded in the generation metadata (details withheld)."
}
],
"run_at": "2026-08-30T22:55:47Z",
"metrics": [
{
"name": "output_duration",
"unit": "s",
"value": 20.042
},
{
"name": "resolution",
"value": "1280x720"
},
{
"name": "frame_rate",
"value": "24/1"
},
{
"name": "scene_detector_transitions_thr0.30",
"value": 2
},
{
"name": "black_intervals",
"value": "none"
}
],
"procedure": "Seedance 2.5, one linked image input (a style reference; the prompt asks to take only its medium, palette and light), a long structured prompt with one scene in four parts (a wide of about 8 s, a close-up of each of two characters of about 4 s requesting spoken lines, and a closing two-shot) and a stated look target (a cobalt night painting look), 20 s, 16:9, audio requested, no seed set. One of a set of sequence runs on 30 August 2026 (UTC). Whether the requested lines were spoken was not checked (audio not listened to). Sampled inspection: 12 timestamped frames from 0.84 s to 19.21 s of a 20.04 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": 20
},
"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_01M64XMY75MG975P7949CPTK86"
]
}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 rail and a lantern keep the terrace readable in blue; the close-up takes a shape of its own.
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. |