Where did this picture come from?
You made 154 pictures for one shot. A week later, someone else opens the one you kept and asks how you got it. Right now the only answer is to ask you.
This tool makes your generative AI work portable and unpackable. It should take about 30 seconds, and it should let them pick up where you left off without starting over.
Six words, so the rest of this page makes sense
I use these words throughout. Here is what each one means in plain terms.
Everything that goes into one press of the button. The words you wrote, the pictures you fed in, which model, which settings. Change any one of those and it is a different recipe.
You press go once. That is one attempt. The provider charges you for that press and hands back whatever it hands back, sometimes one picture, sometimes sixteen. Your 4-press try that returned 16 pictures cost $0.32.
It gets a permanent ID. When one press hands back four pictures, the ID is how you point at this exact one.
An input to the model: a style you want copied, a character who has to stay the same, a colour palette. Order matters to some models.
For any picture: which words made it, which references went in, which model, who ran it, when. When the trail is broken the page says so instead of guessing.
You see a picture you like and carry on from there. The tool remembers you started from that one, so the path back to it never gets lost.
How far out you can zoomFour zoom levels, and the contact sheet is one of them
Everything below is one shot. You made 154 pictures trying 16 recipes, in about half an hour. A scene has 15 to 30 shots. The contact sheet below is the widest view, and it earns its place: you can see the whole run at a glance and the colour shifts alone tell you where the work turned. What the tool will not do is scatter those same pictures as loose boxes and arrows, because at this count that stops being a map and becomes a ball of wool. The map that does work is on screen three, and it groups recipes by the ingredients that went into them.
Screen one: all your shotsWhich shots exist, and how each one is doing
One card per shot. The picture on the card is the one you kept. If you have not kept one yet, the card says that instead of showing you a random pick.
no generations
Screen two: inside one shotWhat you tried, in what order, and what is still alive
This is the screen you would live in. Time runs left to right. Each column is a burst of tries. Inside a column, recipes that only differ a little get stacked together, so 154 pictures read as about a dozen decisions instead of a wall.
Screen three: the mapWhich recipes share ingredients, and when you tried them
You cannot pick what to compare by looking at thumbnails. Two pictures can look identical and share nothing. Two pictures can look completely different and come from almost the same recipe. So this map places recipes by what went into them. Each row is a set of recipes built on the same style picture. Position along the row is the actual time you ran it.
50 pictures
8 pics
8 pics
8 pics
8 pics
8 pics
8 pics
2 pics
24 pictures
8 pics
8 pics
4 pics
4 pics
16 pictures
16 pics
16 pictures
16 pics
16 pictures
16 pics
32 pictures
16 pics
16 pics
A mode, not a screen: line them up by modelOnly earns its place when you are running more than one model
You flagged that this one is niche, and you are right, so it drops from a top-level screen to a mode you switch on. Each row is one recipe, each column one model, and a dash means you never ran that pairing. On this shot it happens to be worth showing, because you ran four models and three recipes crossed between them.
| recipe (and what went in) | midjourney-v7 | midjourney-v8 | krea-2-large | seedream-5-pro |
|---|---|---|---|---|
| 304b1513fc styl·c1473c same recipe, other model |
— |
|
|
— |
| 4c430985b1 styl·2d9f67 same recipe, other model |
— |
|
|
— |
| 315ca06ab3 styl·2d9f67 char·33a5bb same recipe, other model |
|
— | — |
|
| d046452a33 styl·c1473c pale·406bae char·33a5bb |
|
— | — | — |
| ce7811467e styl·c1473c pale·406bae |
— |
|
— | — |
| aa39be3923 styl·c1473c char·33a5bb |
|
— | — | — |
| 2759f3a1be styl·fe5d5f |
— |
|
— | — |
Screen four: one picture, everything about itThis is your Shot Details, with one amendment
You asked whether this should be what opens when you click Shot Details on screen one. Yes, and here is the wrinkle worth deciding now. A shot holds many pictures, so Shot Details has to pick one. It should pick the one you kept, since that is the shot's answer, and it should carry the shot's totals alongside it: how many recipes, how many pictures, what it cost, who worked on it. That gives you two different clicks on the same card. Clicking the picture takes you into the work on screen two. Clicking Shot Details gives you the record. If nothing has been kept yet, it opens on the totals with the picture slot empty.
char_sheet 1ae6ae0d ─┼─▶ this attempt ─▶ 16 outputs
parent ─ not recorded
| this picture | 461038eca0c1362f |
| this try | 20260806T062412Z_midjourney-v7 |
| this recipe | 252f82b4acdea442 |
| model asked for | midjourney-v7 |
| model that ran | — not written down |
| who ran it | midjourney |
| their job number | — not written down |
| seed | None |
| times pressed | 4 |
| pictures back | 16 |
| cost | $0.32 |
the pictures you fed in, in the order you fed them
f1f082404f4e
1ae6ae0d4a47
Screen five: what changed between these twoThe one nothing else does
Two tries, words that changed on top, pictures underneath. The part you flagged as missing is how you get here. You cannot choose a pair by squinting at thumbnails, so the way in is the map on screen three: find two recipes that share ingredients, then open them here. The map is the selector and this is the reading room.
How you would actually use itFour jobs. The first one is how we judge whether any of this worked
If someone cannot do the first one, nothing else matters.
How the tool decides two recipes are relatedAnd why it will never put a similarity score on your words
The map groups by ingredients because ingredients are written down. Words are harder, and getting this wrong would quietly teach you something false.
Measuring how much text changed is the wrong measurement
Take your two recipes from the compare screen. They differ in medium, colour grade, film stock, lens, lighting and style picture. Compared as whole blocks of text they score 0.893 similar, which reads as nearly identical. Scoped to the one section that actually moved, they score 0.355. The template swamps the signal.
One word can carry the whole shot
Swap black for asian and a different person is standing in the frame. Swap fast for slow motion and the shot reads differently end to end. Off-the-shelf text-similarity models score those swaps as almost no change at all, because they are built to treat that kind of substitution as a rephrasing. They are answering a question nobody here asked.
So the rule is: show the words that changed
A number like 0.94 sitting beside two recipes tells you they are basically the same. Sometimes they are the same, and sometimes one word turned the whole shot over. You can read black → asian and know immediately which case you are in. The tool shows you the change and leaves the judgement with you.
The question actually worth asking
Small change, big difference in the picture. That pair is the most useful thing in your whole archive, because everything else stayed still by accident, which is as close to a fair test as real work ever gets. It is also the pair a similarity ranking would bury, because it looks like a near-duplicate.
How this feeds back into making thingsThe part that makes it worth building at all
You said this is not urgent, so here is the thinking rather than a spec. The point of the whole thing is that someone opens your work, finds a picture that landed, and starts from its recipe. That has three consequences and they all point the same way.
The recipe has to travel without the picture
A kept picture is a starting point, so its recipe has to come away clean and run again on its own. That is what the Remix button is. It also means the map should filter down to kept recipes only, which for someone arriving cold is probably the most useful view in the tool: here are the doors you can walk through.
The ingredients have to still be there
A recipe that cannot run is a museum piece. Your style pictures currently live at file paths and temporary web links, and temporary links expire. A shelf of starting points that fail when you press go would be worse than having no shelf, so keeping a permanent copy of every ingredient stops being tidiness and becomes the thing that makes branching possible.
New work lands next to what it came from
When someone branches from a recipe, their attempt inherits most of its ingredients, so on the map it appears right beside the thing it grew from. The map thickens around the work that succeeded. That is the tree you described, and it only draws itself once the tool writes down what each try branched from.
Labels the tool has to carryEach one changes what the tool is allowed to tell you
These are not decoration. Get one wrong and the tool starts making promises it cannot keep.
| label | what it can say | why it has to exist |
|---|---|---|
| how much trail | all of it · some of it · none of it | All 16 of your real ones are some of it. A label is cheaper than a wrong answer. |
| can this model repeat itself | no · takes a seed · tells you the seed · usually repeats · proven to repeat | A plain yes-or-no already gets Midjourney wrong. This decides whether the compare screen is allowed to offer you a real test. |
| how you carried on | ran it again · copied the recipe · used the picture · combined two · started fresh | Copying a recipe and feeding in the picture are different things. Lumping them together hides that an input changed. |
| why a strand stopped | you dropped it · the round ended · it has gone quiet | A strand with nothing after it might be rejected, unfinished, or just resting. Three different meanings. |
| what a try is doing | queued · running · finished · failed | Where it came from is written down before it runs, so a crash leaves a try you can recover instead of a picture with no history. |
Four things I am refusing to buildAnd why each one would hurt you
These are the real decisions in this proposal. Everything else is layout.
1. No loose scatter of every single picture
I had this one too broad and you were right to push. A map is fine when distance means something recorded. Screen three is a map, and it works because the rows are ingredients written down in the receipts. What stays banned is the other thing: 154 individual pictures floating in one space, positioned by a physics simulation. That looks authoritative while being unreadable, which is the worst combination.
2. Nothing sits near anything else by accident
You already have a 3-D graph that drew five tidy clusters. They looked like taste. They were just which board the pictures came from, with nicer names. Two pictures sat together because of where they were saved, not because they looked alike. When a layout pushes things together, people read meaning into it.
3. No leaderboard of which reference works best
You would only be counting the ideas you chose to keep pushing. You stop early on the ones that look weak, and you change the words at the same time as the pictures. A ranking built on that is wrong, and it is convincing, which is the bad combination. The tool reports counts with their totals beside them and leaves ranking alone.
4. No similarity score on your words
The tool shows which words changed and stops there. A score would put a one-word swap and a full rewrite on the same scale, and those two behave completely differently. Section 09 has the numbers from your own recipes.
The lookTwo options, both already yours
I used the cream one, because this screen would sit right next to the tool you already use every day and it may as well match. The dark one is here so you can tell me I picked wrong.
purple = where you are
green = done
these are the exact colours your
package tool already uses
The purple and green already mean the right things in your other screens, so nothing new to learn.
gold marks the thing that matters
this is your idea-harvest page,
built for picking through
thousands of things
Honestly a fair claim, since it was built for exactly this kind of sifting. It just belongs to a different project.
Five questions for youYou can answer these without reading anything above
Which look do you want?
Cream page with black text, like the tool you already use. Or dark grey with gold, like your idea-harvest page. I built the cream one. Either is a day of work, so pick on taste, not effort.
What should be on screen when you open it?
All your shots as cards, so you choose where to go. Or straight into the shot you were last working on. I guessed the second one, because you tend to sit inside a single shot for hours. If that is wrong, say so.
Should the tool guess when a new round starts, or should you tell it?
Right now it guesses by looking for gaps of more than two minutes, and it admits it is guessing. The other way is a “new round” button you hit before you start. One extra click buys you a fact.
The compare screen: early or late?
It is the only thing here nobody else has, which argues for building it first. It is also the one that can confidently show you a pattern that is not real, which argues for building it last, once the record underneath is solid.
Lots of small pictures, or fewer big ones?
This page is packed on purpose, so you can see a whole shot at once. The trade is that every picture is small. If you would rather see six big ones than sixty small ones, that is a different tool and now is the time to say.