Twelve wet-stroke runs, one or two linked images: the palette follows the prompt
Twelve 31 August runs share a wet-stroke look target and one or two linked images; each shows the palette its prompt states, and four are flagged for a glossy render or a photographic face.
Twelve Seedance 2.5 runs of 31 August 2026 (UTC) use a wet-stroke look target, a different scene and stated palette in each prompt, and one linked image (five runs) or two linked images (seven runs) as a style reference. H01 viewed 12 timestamped frames of each, about 1.7 s apart.
Seen in the sampled frames: the palettes are amber, red, brass-amber (in two), ice cyan, jade green, silver with one red seam, gold, white-hot, verdigris, vermilion and violet, each with true black; each follows the prompt's stated palette. Eleven of the twelve follow a wide to about 7.5 s, close-ups at about 9.2 to 14.2 s and a closing frame from about 15.9 s; one reaches its first close-up at about 7.5 s. In the seven two-image runs no obvious copy of the second image's composition is seen at this size (this is not a pixel comparison).
Request versus picture, at the sampled size and as H01's judgement: four runs read as glossy renders with few visible brush marks, and in one of them a woman's close-up also reads as photographic skin, though the prompts ask for broad wet brushstrokes and no photographic skin.
Caution: the sampling interval is 1.7 s, so moves between samples are not seen; twelve single runs (no seed set); each linked image has an unconfirmed disclosure and rights basis and none is reproduced; whether requested spoken lines were produced is unknown because audio was not listened to.
Reading: with one or two reference images, the palette that the prompt states wins in these runs; this describes these twelve runs, not a rule.
These are retrospective observations of founder tests, not a benchmark or product validation.
Structured fields
{
"citations": [
{
"note": "https://veyrul.com/experiments/exp_01M64VFHPMBN34CE0V0A7FYC4W (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFQZ7K0W809MVJGFRNFMK (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFW041QDQDACG4KZ91BYK (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFFMG8GMKQDKKWMKN0CXQ (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFMRY8F46PSBB2ABE3XZS (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFG92WN7ATN99ASSHR1ZF (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFNC4EG9SA11E4YYZQ2FQ (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFVA9V5H2QHA5GDFMH79X (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFPQ38TJ63ZYG803FJH8N (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFEZYFGJ8SAZGY78BX5VP (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFGWWSGKBK3WQ3T3J1GSZ (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64VFJW4E7R565ZKZQPEZS0R (candidate; Main resolves to a record ID)"
}
]
}Evidence claims
- [inference] Synthesis of H01's sampled-frame observations of the cited candidate experiments.
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. |