Is it one person's face from shot to shot?

We have not measured it, so we do not claim it. What we measured and can say held is the wardrobe, the hair, the props and the set.

How does a character keep its wardrobe?

Its main reference photo rides as a reference image with every prompt it is attached to, and its description rides in as words. Both say the same jacket, boots and bob, and the model renders them. The description can also bring things the prompt did not ask for: the umbrella in shot 3 came from it.

Why did the case vanish from one shot?

Because its sentence was not in that frame's prompt. A prop is only as consistent as the words that describe it in each prompt, so paste the same sentence into every one; the second version of that frame, with the sentence, has the case.

Does the location still have to be a photo?

No. Ours is a still made with Seedream 5.0 Pro. Any still of the place, uploaded or generated, works as the location's reference.

Can I attach the character straight to the video model?

On the models we tried, not well: Seedance 2.0 Fast refused her reference, Seedance 1.5 Pro cannot carry one, and Seedance 2.5 took her but made up its own platform. A 16:9 frame from the image composer with the location attached, then Animate an image, carried both.

How was the palette measured?

Every frame at two frames per second, the share of saturated pixels in three hue bands, averaged per shot, against the same measure on the location's still.

Guides / Filmmaking

Keep characters and locations consistent across shots

Consistency on NOLGIA is two things: reference images, and the same words every time. This guide takes the five shots of Last Train and checks what stayed put: the wardrobe, the props, the set and the palette. It counts. It does not claim the face belongs to one person, because we have not measured that.

11 min read
September 24, 2026

Anchors

Shots

The count

What you make, in order. Tap one to jump to its step.

In short

A character is reference images plus a description; attach it and write @Name in the prompt. A location is the same for a place. A prop is a still plus one sentence, pasted into every prompt.
Across five shots, every wardrobe item in frame held in every shot with Vale in it, and all six set elements held in all five shots. The case held exactly where its sentence was in the prompt and vanished from the one frame where it was not; the umbrella, which is in Vale's saved description, turned up in a shot whose prompt did not name it.
The palette held where the frame shows the set: the run, mostly wall, is within a few points of the still's green; shots filled by a train or by Vale carry less green and more amber or red, as they should.
The face is not on this list. Nothing here measures it, so nothing here claims it.

The shots in this guide are the five shots of Last Train, a 25 second film made for these guides on the company's own account. Vale is a saved character, Halcyon Street station is a saved location, the chrome case is a still and a sentence. Every shot was made as a 16:9 frame in the image composer with the character and the location attached, then animated in the video composer. Every person in it is fictional.

We then went through the shots with a list and a measuring tape: which garments, which props and which parts of the set appear in each one, and how much of each frame is the set's green, the tunnel's amber and the jacket's red. What follows is that count, including the two places it fails.

What goes wrong

Mistake	Why it fails	Fix
Describing the character afresh in each prompt	Each prompt is its own request; a coat becomes a jacket, a bob becomes a ponytail	Save a character with a reference image and one description, attach it, write @Name
Describing the place in one shot and assuming it	The model invents a new platform every time	Save a location with a still and one sentence; attach it to every shot
Leaving a prop out of one prompt	The prop is gone from that shot, whatever the last shot showed	One sentence per prop, pasted into every prompt where it should appear
Sending the character straight to a video model	One model refuses a photoreal reference, another cannot carry one, and the one that takes her invents its own set	Make a 16:9 frame first with the cast attached, then animate it
Trusting a good first second	Sets repeat and props migrate as a clip runs on	Keep shots at 4 to 6 seconds and cut where it drifts

What an anchor is here

There is no trained identity on NOLGIA. There are three kinds of anchor, and this guide uses all three:

A character, from the Characters page: up to eight reference photos with a main one, and a description that rides into every prompt it is attached to. Pick it under Characters in either composer and write @Vale in the prompt. The composer puts the main photo in a reference slot and rewrites @Vale to that slot.
A location, from the Locations page: a still and a canonical description. Pick it under Location and the still rides as a reference with the sentence.
A sentence, for anything else. The chrome case is one sentence, pasted unchanged into every prompt it belongs in, plus its own still as a reference image for the insert.

Vale's reference portrait: platinum bob, cherry-red leather bomber, white tank, wide black jeans, platform boots, clear umbrella. — **The character's main reference.** Five wardrobe items and one prop to check for.

An empty underground platform with pale green tiles, a dark green stripe, fluorescent tubes, a steel bench and a warm amber tunnel mouth. — **The location's reference.** Six set elements to check for.

A brushed-chrome hard-shell case with a black handle and two black latches. — **The prop's still**, and its sentence: a small brushed-chrome hard-shell case with a black rubber handle and two black latches.

Vale's description, as saved on the character

The lead of Last Train: a woman in her early twenties with a platinum blonde sleek chin-length bob and a straight fringe, an oversized cherry-red leather bomber jacket over a white ribbed tank, wide black jeans, chunky black platform lace-up boots, a folded transparent bubble umbrella in one hand.

The five shots

Each shot below was a frame first, made in the image composer on GPT Image 2.5 Flare with Vale and the station attached, then animated from that frame: the run on Kling v3, the other four in the video composer on Seedance 2.5. The insert's frame had no character: the station attached and the case by its sentence.

Vale running toward the camera down the wet platform, the chrome case in her right hand and the umbrella in her left. — **Shot 1, the run.** Kling v3.

Vale passing the busker, Ozzy, with the chrome case and the umbrella; he holds his saxophone and watches her go. — **Shot 2, the busker.**

Vale reaching for the doors of a train as they close, the train pulling away. — **Shot 3, the doors.**

The chrome case standing alone on the wet platform, a train streaking past behind the case. — **Shot 4, the insert.**

Vale sitting on the steel bench, out of breath, smiling as the camera closes in. — **Shot 5, the bench.**

What held: wardrobe, props and set, shot by shot

We looked at every shot two frames per second and ticked a list: five wardrobe items (the cherry-red leather bomber, the white ribbed tank, the wide black jeans, the black platform boots, the platinum bob), two props (the clear umbrella, the chrome case), and six set elements (pale green tiles, the dark green stripe, the fluorescent tubes, the steel bench, the yellow line, the amber tunnel mouth).

Shot	Wardrobe	Umbrella	Case	Set
1 The run	5 of 5, jacket open	Yes	Yes	6 of 6
2 The busker	5 of 5, plus Ozzy's windbreaker, hoodie and sax	Yes	Yes	6 of 6
3 The doors	4 of 5: the boots are out of frame	Yes, though not in this prompt	Yes	6 of 6, plus the train
4 The insert	No person	No	Yes, still from the first frame	6 of 6, plus the train
5 The bench	5 of 5	Yes	No: set down in shot 4	6 of 6

Every wardrobe item in frame, in every shot with Vale in it; every set element in every shot. The case appears exactly where its sentence was in the prompt. The umbrella also appeared in one shot whose prompt did not name it, carried in by Vale's saved description.

Watch for this:The gap: the case vanished from the first frame 2. Our first version of frame 2 named Ozzy, the bench, the sax and the umbrella, and not the case. So it had no case, although she carries it in frames 1 and 3. Nothing carries a prop forward for you. The fix was the boring one: the case's sentence added to the prompt, and the frame made again.

Frame 2, first version: Vale walking past Ozzy and his saxophone with only the umbrella in her hand; no case — The same character, the same location, one sentence added.

Frame 2, second version: Vale walking past Ozzy with the chrome case in her right hand and the umbrella in her left — The same character, the same location, one sentence added.

Watch for this:The reverse: a prop the prompt never named. The reverse also happens. Frame 3's prompt names the case and not the umbrella, and the umbrella arrived anyway, because Vale's saved description puts one in her hand. A character's description rides into every shot it is attached to.

What held: the palette

A set is also a color. We measured every frame of every shot at two frames per second: the share of saturated pixels whose hue falls in the set's green band, in the tunnel's amber band, and in the jacket's red band, and the mean brightness. The location's still is the reference row.

Shot	Green	Amber	Jacket red	Brightness
Location still (reference)	69%	1%	0%	0.26
1 The run	65%	3%	3%	0.28
2 The busker	33%	6%	9%	0.28
3 The doors	19%	12%	8%	0.25
4 The insert	37%	17%	0%	0.35
5 The bench	45%	13%	11%	0.28

Share of each shot's pixels in each hue band, saturated pixels only, averaged over its frames at two frames per second. The set's green runs from 19 to 65 percent against the still's 69: it follows how much of the frame is tiled wall. Amber rises from the still's 1 percent wherever the tunnel glow, a train's lights or the warm haze come into frame. The jacket's red tracks how much of the frame Vale fills, and is absent from the insert, as it should be.

The run is closest to the still: a wide shot down the platform is mostly wall. The doors shot has the least green because the train fills the right of the frame, and the insert is the brightest because a train's headlights come out of the tunnel behind the case. Neither is the palette drifting; both are what the shot shows.

What did not hold, and what we do not claim

The face. We did not measure whether Vale's face belongs to one person from shot to shot, so this guide does not say it is. The wardrobe, hair and props are what the count above covers.
A prop without its sentence. The case was missing from the first frame 2 because its sentence was not in that prompt.
A prop from the description. The umbrella arrived in shot 3 without being in the prompt, because Vale's saved description puts it in her hand. Write the shot around the description, or say what each hand holds.
One take each, this time. None of the five shots needed a second take. On another roll a prop can still migrate or fade in, so check every take at full size.

Vale on a bench on a pillared platform, made with no frame and no location attached. — **Why frames first.** Attached straight to Seedance 2.5 with no frame, Vale arrives, on a platform the model made up: pillars, no green stripe, no amber tunnel. Seedance 2.0 Fast refused her reference, and Seedance 1.5 Pro cannot carry one. A 16:9 frame with the location attached carries the set; the direct route does not.

Which anchor for which job

You want to keep	Use	What it holds here
A person's wardrobe and hair across shots	A character, attached, @Name in the prompt	Every item in frame, in 4 of 4 shots with her in them (the boots out of frame in one)
A place across shots	A location, attached	6 of 6 elements in 5 of 5 shots; the set's green in every shot, its share following how much wall is in frame
A prop across shots	One sentence in every prompt, and a frame with the prop in it for a static insert	Present wherever the sentence was, absent from the one frame where it was not
Two people in one shot	Both in the cast, in order, named with @ in the prompt	Vale and Ozzy, shot 2
A face	Nothing measured here	Not claimed

What it costs

Saving a character or a location costs nothing. A frame is priced per image, a shot by model and length. Live rates for the models used in these five shots:

GPT Image 2.5 FlareEvery plan
Per image
Native 8 credits · 2K 21 credits · 4K 58 credits
Kling v3Pro and up
Per 5s clip (audio on by default)
720p 35 credits · 1080p 47 credits · 4K 117 credits
Seedance 2.5Pro and up
Per 5s clip
480p 26 credits · 720p 56 credits · 1080p 137 credits
1080p renders at 16:9 and 9:16
Seedream 5.0 ProEvery plan
Per image
1K 5 credits · 2K 9 credits

Read live from the catalog when this page loads. Every price is shown before you generate. See every rate.

Checklist

One character per recurring person, saved with a main reference and one description.
One location per place, saved with a still and one sentence.
One sentence per prop, in every prompt where the prop should be seen.
Frames at 16:9 with the cast and the location attached; shots animated from the frames.
Every shot checked against the list at full size before the cut; a gap is a redo of one prompt, not the whole film.

Try it yourself

Save the portrait below as a character's reference and the station as a location's reference, attach both in the image composer at 16:9, and use frame 1's prompt from the short film guide. Then run your own count: five wardrobe items, two props, six set elements, shot by shot.

The anchors

Vale, the reference portraitWebP image · 1536x2048 · 238 KBDownload
Halcyon Street station, the reference stillWebP image · 1536x864 · 114 KBDownload
The chrome case, the prop stillWebP image · 1280x1280 · 56 KBDownload

Questions and answers

Is it one person's face from shot to shot?: We have not measured it, so we do not claim it. What we measured and can say held is the wardrobe, the hair, the props and the set.
How does a character keep its wardrobe?: Its main reference photo rides as a reference image with every prompt it is attached to, and its description rides in as words. Both say the same jacket, boots and bob, and the model renders them. The description can also bring things the prompt did not ask for: the umbrella in shot 3 came from it.
Why did the case vanish from one shot?: Because its sentence was not in that frame's prompt. A prop is only as consistent as the words that describe it in each prompt, so paste the same sentence into every one; the second version of that frame, with the sentence, has the case.
Does the location still have to be a photo?: No. Ours is a still made with Seedream 5.0 Pro. Any still of the place, uploaded or generated, works as the location's reference.
Can I attach the character straight to the video model?: On the models we tried, not well: Seedance 2.0 Fast refused her reference, Seedance 1.5 Pro cannot carry one, and Seedance 2.5 took her but made up its own platform. A 16:9 frame from the image composer with the location attached, then Animate an image, carried both.
How was the palette measured?: Every frame at two frames per second, the share of saturated pixels in three hue bands, averaged per shot, against the same measure on the location's still.

Save the cast and the set once

Open Characters

Filmmaking

A short film with AI: the pipeline from script to final cut

A 25 second film, Last Train, from a five line script to a graded cut: a saved character, a saved location, a prop, frames first, then shots, sound, Studio, and the export to Premiere or Resolve.

September 24, 2026 · 13 min read

Video editing

Put your own character and your dog into any video

Pick a video, add a photo of a person and one of a dog, and Miroge recasts the clip with them: the same moves, the same set, the same camera. Every step, on a computer and a phone.

September 23, 2026 · 10 min read