Presentations: AI Features

Use AI to generate entire decks, refine individual slides, create backgrounds, and apply visual styles, all from natural language prompts.

Chat Mode Beta

Chat Mode is a split view that turns the presentation editor into a research workspace. A chat panel sits on the left and a live slide preview on the right — ask Theo to research the web, rewrite slides, or edit imagery, and the deck updates beside you in real time.

  • Open it from the top bar: click the Chat pill in the editor’s top bar, right next to Theo (on a phone it lives in the bottom toolbar). The Chat Mode card on the settings panel’s Theo tab opens the same view, and on the Presentations page any deck’s menu has Chat with Theo, which opens that deck straight into Chat Mode. Click Back to canvas to return to the full editor at any time — every change Theo makes stays undoable there, and the view you left is remembered when you come back.
  • Theo sees your slides: when you talk about a slide — “this slide”, “slide 3”, “does the layout look crowded?”, or with an area selected — Theo looks at that slide’s current image, not just its text. It reads the words actually on the slide and judges the real layout, contrast and legibility, so the answer is about what you see, not a guess.
  • Ask for a review: say “review this slide” or “what’s wrong here?” and Theo returns a review card — a score out of 100, a one-line verdict, what is working, what is missing, and a few concrete suggestions. Each suggestion is a chip: tap one and Theo makes that change. Ask to “review the deck” and you get every slide’s score plus the two or three places to start. Nothing changes until you tap.
  • Web research: Theo searches the live web for current facts and statistics, then shows the sources it used right in the thread so you can verify and cite them.
  • What it can edit: a slide’s title and body copy, speaker notes, the order of the deck (add, remove or move slides), a chart’s data, and — on Creator slides — the slide image itself: change one part of it, replace the exact words baked into it, work an attached photo or logo into it, or rebuild it from scratch. Just describe the change.
  • Theo tells you the plan first: before a slide image changes, Theo says in one line what it is about to do — “Replacing the title with ‘Q4 Results’”, “Swapping the person on the right” — and changes only that. Everything else on the slide stays exactly as it was, including your earlier edits.
  • Theo asks when unsure: if your request could mean two things (“make the logo bigger” on a slide with two logos), Theo asks one question with a few options instead of guessing. Pick one and it carries on; nothing is rendered until you do.
  • Edit imagery in place: attach a character, logo, or photo and ask Theo to incorporate it, or change one part of a slide (“make the graphic on the left 3D”). Theo edits the image that is on the slide right now and keeps the title, bullets, logos, and layout intact — it never wipes the slide to drop in a bare image, and a single attached image is used on the slide, not as the slide, unless you say so.
  • Every edit is a version: image edits made in Chat Mode land in the slide’s Image versions just like edits from the canvas, with the same Compare and Undo — and Theo remembers them, so “make the person I added a bit smaller” needs no restating.
  • New imagery from scratch: ask Theo to build a brand-new slide image. It uses your deck’s own engine by default — ask for a premium image to push it to Theo Ultra, or a fast one to drop to Theo Creative. You can also surgically edit a selected region, or use Theo Isolate to cut a subject onto a transparent background.
Chat Mode is on for every deck and focuses on conversational research, review and editing. For pixel-level layout work (dragging elements, fine-tuning positions), switch back to the canvas editor.

Chat History & Resuming a Conversation

Chat Mode conversations are saved per presentation. When you leave and come back — or exit Chat Mode and re-enter it — your last conversation is restored automatically, so you never lose the thread you were working in.

  • History menu: the chat icon at the top of the chat panel opens a dropdown of every conversation for this deck (with a count). Pick one to switch to it — the thread and its edit cards load back in.
  • New chat: the + button starts a fresh conversation while keeping your previous chats in history. A new chat is only saved once you send the first message.
  • Auto-titled: each conversation is named from its first message so it’s easy to find later.
History is scoped to each presentation, so a deck’s chats stay with that deck. Switching chats mid-stream safely stops the in-flight turn before loading the other conversation.

Version History & Restore

Chat Mode keeps a durable version history of your deck so you can always go back. Open it from the History button in the preview toolbar (top-right).

  • Automatic checkpoints: a version is saved automatically right before each AI edit, so any change Theo makes can be rolled back — even after you’ve moved on. (Pure research / Q&A turns don’t create a version.)
  • Save a version anytime: click Save version in the panel to bookmark the current deck before a big change.
  • Preview before / after: click Preview on any version to see it side-by-side against your current deck, with synced slide navigation and a per-slide Changed / Added / Removed marker.
  • Restore: bring a version back with one click. Restoring first snapshots your current deck (so the restore itself is reversible), then swaps the deck in place — the change autosaves like any other edit.
The most recent versions are kept per deck (older automatic checkpoints are pruned over time). Restoring never loses your current work — it’s saved as a “Before restore” version first.

AI Wizard: Generate a Deck

Open Presentations → New and the wizard walks you through four steps — Describe, Focus, Look, and Outline — on one calm, paper-white surface with a step track at the top so you always know where you are.

  1. Describe: write your brief in natural language. A sentence is enough, and a full brief is better — there is no practical length limit, so paste the whole thing. Theo reads all of it. Attach a project or sharpen the text with Enhance from the row under the brief, pick a deck length (or let Theo decide), or hand Theo your website and let him craft the brief for you (see below). When you’re ready, the Continue pill floats at the bottom of the step — it tells you what comes next, and /Ctrl + Enter inside the brief does the same.
  2. Focus: tell Theo what kind of deck this is, who it’s for, and the tone. When Theo crafted the brief from your site, this step arrives pre-filled — change anything.
  3. Look: choose a theme or describe your own (e.g. "dark, corporate, blue accents"). If Theo read your website, a brand-matched theme is suggested first.
  4. Outline: review and edit every slide card, pick the slide engine, refine with an instruction, then generate. Theo writes titles, body text, speaker notes, and suggested layouts for every slide.
Your draft — brief, sources, focus, and theme — is saved as you type, so a refresh never loses your place. You can also trigger the wizard from an existing project by opening the AI Sidebar and choosing "Regenerate All Slides."

Craft with Theo & the Context Layer: Start From Your Website

On the wizard’s Create tab (the Describe step), the Craft with Theo row turns a website and a few rough ideas into a complete brief. Paste your site, add a line about the deck you want (“investor update for our Series A, confident but not salesy”), and click Craft my brief — Theo reads the site and writes the whole brief for you.

  • The brief is yours: it lands in the brief field as editable text. Trim it, sharpen it, or paste more — whatever you send forward is what Theo builds from.
  • Context layer, filled in: everything Theo learned from the site becomes an editable Context layer card under your brief, with a source chip linking back to the page.
  • Focus, pre-filled: the deck type, audience, tone, and company name Theo inferred arrive pre-filled on the Focus step — every field is still yours to change.
  • A brand-matched look: your site’s colors and fonts are turned into a suggested theme on the Look step. Keep it or pick another.
  • Ideas only: no website? Write your rough ideas and craft anyway — Theo expands them into a full brief. And if a site can’t be read, Theo still writes the brief from your notes and tells you so.
  • Context only: already have a brief? Choose Add as context only to extract the page into a Context layer without touching what you wrote.
  • Feeds the outline: when you generate, the outline is built from your brief plus the Context layer, so slides ground themselves in real facts instead of inventing details.
  • Replace or remove anytime: craft again from a different URL, or remove the Context layer to go back to a brief-only generation.
Theo distills the page into an overview, what they do, key facts, and tone — and only keeps what the page actually says. Both the brief and the Context layer are saved with the wizard draft, so they survive a refresh.

Import & Merge: One Flow, With Direction

The wizard’s Import tab handles both single-document imports and multi-source merges in one flow. Stage 1–5 sources (PDFs, PowerPoints, or existing OpenCharts decks) and use the “Tell Theo what to do with it” field to give the import a direction — condense into an exec summary, translate to another language, refocus on the financials, or modernize the structure. Leave it blank for a faithful recreation.

  1. Open Presentations → New and switch the segmented control to Import.
  2. Drop a file (or add sources with Upload PDF or PowerPoint / Add an existing deck).
  3. Write your direction (e.g. “Condense this into a sharp 10-slide executive summary” or “Merge into a 12-slide investor pitch keeping Deck A’s visuals”).
  4. Click Continue. One source flows through the theme step (with automatic brand/theme extraction for files); two or more sources merge and jump straight to the outline step.

Merges recompose 2–5 sources into a single, brand-new presentation — never a naive concatenation. Theo synthesizes a fresh Hook → Context → Evidence → Insight → Close narrative across every source, dedupes overlapping slides, and writes the merged deck around your instruction.

Visual-style choices (2+ sources)

  • First source (default): use the first source’s theme, fonts, and visual style.
  • Specific source: click the Visuals radio next to whichever deck’s look should win.
  • Blend: mix the first two themed sources’ palettes — primary from source 1, accent + body font from source 2.
  • Fresh look: ignore source themes and start with the default OpenCharts theme.

Theo Chat shortcut

In any chat conversation, attach 2 or more presentation projects and say something like “merge these into a single deck”. Theo will call themerge_presentationstool with the same non-concatenation contract and ship the new deck with a “Merged from N decks” card.

The non-concatenation contract is hard-wired server-side: Theo recomposes 2–5 sources into a single narrative, never Source A’s slides + Source B’s slides. If the result drifts from your instruction, refine the outline in step 3 or click Regenerate outline.
Merge is built for resilience — Theo automatically falls back to a different engine if the primary takes too long or is briefly unavailable, so you almost never see a hard failure. Just retry once if you do.

Adding a Slide (Creator Mode)

Clicking the + Add slide button in the thumbnail panel opens a focused modal where you decide exactly what the next slide should be before it lands on the deck.

  • Theo, suggest the next slide: a Sparkles button that reads your existing slides (titles, bullets, speaker notes, and the original deck prompt) and proposes a natural follow-up concept. The suggestion drops into the textarea so you can keep editing it before submitting.
  • Concept presets: one-click chips (Section divider, Key statistic, Process step, Closing thought) that seed the textarea with a starter prompt.
  • Reference image (Creator mode): drop, paste, or upload an image. Theo classifies it and lets you choose whether to use it as the slide image directly or as a style reference threaded through the anchored cascade. Logos and icons default to the style-reference path so the asset never overwrites the whole slide.
  • Visual style override: a condensed style picker defaulting to the deck or active slide's style. Pick a different look for just this one slide without changing the deck-wide style.
  • Add blank slide: the escape hatch in the modal footer skips the prompt entirely and inserts an empty slide the same way the old flow did.

Once you submit, the new slide is inserted after the current one and — in Creator mode — immediately enters the existing rebuild shimmer until the image is ready. Keyboard shortcuts: ⌘ ↵ to add, Esc to close.

The AI-suggest button uses Gemini Flash for low-cost suggestions — well under a credit per click. If you run out of credits, the textarea stays editable so you can keep typing manually.

Per-Slide AI Prompt Bar

Each slide has a prompt bar at the bottom. Type a command to modify just that slide:

  • "Add a chart showing revenue growth"
  • "Make this slide more visual"
  • "Simplify the bullet points"
  • "Change the layout to two-column"

The AI analyzes the current slide content and applies targeted changes.

Attaching Images: Upload, Paste, Drag-Drop, Multi-Reference

The per-slide AI bar accepts up to 12 reference images per turn, via any of three input methods:

  • Upload: click the image button to open a file picker. Select multiple files at once — they all attach as separate references.
  • Paste:⌘V/Ctrl+Vwhile the bar is visible. Works whether your cursor is inside the text input or anywhere else on the page.
  • Drag-and-drop: drop image files onto the bar. A pink dashed overlay confirms the drop target.

Each newly attached image is analysed in parallel by a lightweight vision pre-flight so the bar can show contextual presets (Swap subject, Match style, Set as logo, etc.) tuned to what you actually uploaded.

Multi-reference editing (Creator mode): with 2 or more references attached, the AI never falls back to the "replace the whole slide image with a single upload" shortcut. Instead, it sends every reference image as labeled bytes (A, B, C, …) to the image model alongside your instruction, so you can swap multiple subjects in one shot:

  • "Swap the three team members with these three reference photos, left to right."
  • "Replace the woman on the right and the man on the left with these."
  • "Use these images as combined style references."

One image is a reference by default: on a Creator slide a single attached image is used on the slide, never silently made into the slide. Phrasing like "swap the guy on the right with this", "use this image" or no prompt at all all route the upload through the reference path. To actually swap the whole slide image, say so ("use this image as the slide") — or flip the chip.

The action chip: with one image attached, a chip in the composer’s toolbar says what Send will do — Use as reference, Replace slide image, Add as element or Set as logo. It follows your wording as you type; click it to choose yourself. What it says is what happens.

Recent references: images you used on this slide before appear as small tiles above the input — one click re-attaches one, so "the mascot again" never means finding the file again.

References larger than 8 MB each — or that push the total payload past 32 MB combined with the slide image — are skipped server-side. The editor logs which ones were dropped so you can resize and re-attach if needed.

Editing a Slide Image Again and Again

Every image edit on a Creator slide — from the composer, from Visual Edit, or from Chat Mode — works on the image that is on the slide right now. Ask for a change, then another, then another: each one keeps everything the earlier ones did. Theo is also told what those earlier edits were, so "make the person I added a little smaller" resolves without restating anything.

  • Theo reads the slide first: before anything changes, Theo works out where every piece of text and every subject sits on the image and turns your request into a plan. The status line says what that plan is — “Replacing the title with ‘Q4 Results’”, “Removing the chart on the left” — and the pill that appears when the edit lands says the same, so you always know what was touched.
  • Only what you asked for changes: a request about one thing edits one area; the rest of the slide is left untouched, pixel for pixel. Ask for a new look across the whole slide and Theo restyles the whole image while keeping the words, the logo and the layout in place.
  • Exact text edits: “Change the title to Q4 Results” finds that exact text on the image, replaces only those words — same typeface, size, colour and alignment — and then reads the result back to check it says precisely what you asked. The slide’s stored title and body are updated to match, so the old words never creep back in a later edit.
  • Theo asks when unsure: when a request could point at two things, Theo asks one short question with a few options right in the composer instead of guessing. Tap an option to continue; nothing is rendered until you do.
  • Honest when it cannot: if the new text cannot be rendered cleanly after a second try, Theo says so and leaves the slide as it was — it never ships a misspelled word or a wrong number. Rephrase, shorten the text, or use Visual Edit to point at the exact spot, and try again. When Theo did apply a change but is not fully sure it landed right, the result pill says so and offers Compare.
  • Compare & undo: right after an edit lands, a small pill appears over the slide. Hold Compare to see the version before; click Undo to go back.
  • Image versions: the Slide tab keeps every image this slide has had — original, each edit, uploads, rebuilds, Chat Mode edits — with what made each one. Click any of them to put it back. Restoring is recorded too, so nothing is ever lost.
Rebuilding a slide or uploading your own image starts a fresh picture, so earlier edits are no longer "in" it — but they stay in Image versions in case you want them back. Text edits work best when you quote the new wording exactly (“change the subtitle to ‘Built for teams’”) — Theo uses your words verbatim and never paraphrases them.

Visual Edit: Change One Part of a Slide

Visual Edit (Creator slides) lets you draw a box directly on the slide and change only what is inside it. Click the Visual Edit toggle on the slide action bar — the bar above the slide switches to a mode bar that tells you what to do next, and the cursor becomes a crosshair. Press Esc or Exit to leave.

  • Draw a box: click and drag over the part of the slide you want to change. Drag the box or its handles to adjust it. The edit card opens next to it — it always stays on screen, even near the edges of the slide or in Chat Mode.
  • Describe the change: type an instruction ("change this title to Executive Strategy", "remove the chart") and press Enter. Theo paints only the inside of the box; every pixel outside it is left exactly as it was — and it checks that afterwards rather than assuming it.
  • Exact text inside the box: when your box covers a line of text and you ask for new wording, Theo replaces exactly those words in the same typeface, size and colour, then reads the result back to confirm it matches — and measures the letters too: if the new words came out a different size or a different colour from the original line (grey where the slide uses lavender, say), Theo repaints them once more with the original named. If it cannot render the new text cleanly after that second try, it tells you and leaves the slide untouched instead of shipping a wrong word.
  • The card tells you the plan: while Theo works, the status line says what it is doing with the box (“Editing the selected area”, “Replacing ‘Q3’ with ‘Q4’”), and the result pill repeats it once the edit lands. If the instruction could mean two things, Theo asks a short question first.
  • Attach a reference: click Attach, drop a file onto the card, or paste from your clipboard — up to six. The first one previews inside the box so you can see what lands where, and the Subject · Concept · Style picker says how Theo should read it. Images you used on this slide before wait in a Used on this slide row — one tap stages one again, no re-upload.
  • Or drop it straight on the slide: while Visual Edit is on, drag a file onto the slide itself (or paste with nothing selected). Drop it on your box and it is attached to that selection; drop it anywhere else and Theo draws a box around the spot for you, so “drag the photo onto the person” is one gesture.
  • Theo tells you what it sees: under the attached image a line says what Theo read and what Apply will do — “Looks like a cutout on a transparent background · will place it as-is”, “3 people in this photo · will blend the one in your box” — so nothing lands as a surprise.
  • The action chip: the card shows what Apply will do — Blend into selection (Theo works the reference into the box), Place as-is (your file is dropped in exactly as it is — logos, stamps, icons, cutouts; no AI touches it) or Replace slide image (the file becomes the whole slide). Theo picks a sensible default from your wording and the image — a logo, an icon or a cutout on a transparent background defaults to Place as-is, a photo of several people never becomes the whole slide by accident — and you click the chip to choose.
  • A cutout is placed, never repainted: attach a product, sticker or icon on a transparent background and ask to swap something for it, and Theo clears the box and puts your pixels there — the exact object you attached, not a look-alike. Ask for it to be blended or to match the lighting and Theo paints it in instead.
  • Keep going: after an edit lands, Visual Edit stays on. Draw the next box — it edits the image that is now on the slide.
A box that fully covers a single image or logo element with one file attached swaps that element directly — instant, no credits. And if an attached image can’t be loaded, Theo says so and stops rather than quietly editing without it: remove it, attach it again, and apply.

Per-Slide Chat Memory & What the AI Sees

The slide AI bar keeps an in-memory chat thread per slide. Every turn — your prompt plus a short assistant summary — stays attached to the active slide so the AI inherits the context on the next turn. You never have to restate which subject "the guy on the right" referred to, which reference replaced who, or what visual treatment you applied two turns ago.

  • The thread expands inline above the input — the most recent 3–4 turns are visible at a glance.
  • Click the New thread button on the thread header to clear the slide's memory and start fresh.
  • Each slide has its own thread — switching slides switches threads, so context from slide 3 doesn't contaminate slide 4.
  • Threads live in-memory only and reset on page refresh. They're a working memory for your editing session, not a permanent transcript.

What the AI sees on every turn: beyond your prompt and any attached references, the bar ships a structured slide context block so the model knows the composition it's editing — not just the rendered image:

  • The slide title, bullet points, and speaker notes.
  • The bounding-box geometry of every element on the slide (title, body, charts, images, shapes) so the AI knows what's where.
  • The slide's layout, visual style, and theme colours (primary, accent, background).
  • Lightweight previews of the previous and next slides plus the overall deck purpose, so cross-slide edits stay coherent.
  • A compressed slice of the last ~10 turns of the slide's chat thread, so callbacks like"keep the same treatment, just brighter" resolve correctly.

Image elements have their srcdata stripped before being sent — only geometry and role are forwarded, so the model gets the spatial picture without re-uploading every embedded image on each turn.

Threads reset on page refresh by design — they're a per-session working memory, not a deck-level transcript. If you need a fresh start mid-session (e.g. you moved on to a totally different edit), tap New thread to wipe just the active slide's history.

Global AI Sidebar

Open the AI sidebar from the toolbar for deck-wide operations:

  • Regenerate All Slides: rewrite the entire deck from a new prompt.
  • Add Slides: generate additional slides on a topic and insert them at the end.
  • Restyle: describe a visual style and the AI adjusts colors, fonts, and layouts across the deck.
  • Summarize: condense the deck into fewer slides.

Focus & Restructure

Give your deck a focus so Theo knows what kind of presentation you're building — an investor pitch deck, a sales / proposal deck, a marketing deck, a financial report, and more. Open the Properties panel (the settings icon on the right) and expand the Focus section at the top.

  • Pick a focus type and set the audience and tone. Theo remembers it on the deck.
  • Focus-aware edits: every deck-wide AI edit ("tighten this deck", "make it punchier") now respects the focus, its proven narrative arc, and your audience/tone — so changes stay on-strategy instead of drifting generic.
  • Restructure to this focus: one click rebuilds an existing deck into the chosen type — it reorders, reframes, and retitles your slides into the proven narrative sequence, adds any essential sections that are missing, and trims redundancy. For a pitch deck that means the opinionated, investor-standard structure VCs expect.

Restructure preserves your content — it keeps the facts, data, names, and core ideas you already have and reframes them for their role in the arc; it doesn't invent fake data or throw away your substance. The whole deck is reorganized in one step, so a single⌘Z(undo) reverts the entire restructure. In Creator mode the slide imagery rebuilds to match.

Focus types include Pitch Deck, Sales / Proposal, Marketing, Financial, Educational, and Internal. Each carries a best-practice narrative arc that Theo follows when editing or restructuring. The Pitch Deck arc is the canonical investor sequence — Problem → Market / TAM → Solution → Product → Team → Traction → … → The Ask — with opinionated guidance on what makes each slide strong (bottom-up market sizing, real traction, founder–market fit, a specific ask).

Translate Your Deck

Click the Translate button (the languages icon in the toolbar) to convert the entire presentation into another language. Pick a target language and Theo rewrites every slide's text while keeping the exact same design, layout, colors, fonts, images, and transitions. Because translating the whole deck is a big change, Theo asks you to confirm first — in Creator mode the confirmation reminds you that every slide image will be rebuilt and redesigned with the translated text (which uses AI credits).

Supported languages:

  • English
  • Spanish (Español)
  • French (Français)
  • Italian (Italiano)
  • Russian (Русский)

Translation covers slide titles, body text, bullet lists, quotes, chart labels, speaker notes, and the presentation title. Things that should never change — images, colors, fonts, icons, code blocks, and slide positions — are left exactly as they were.

The whole deck is translated in place as a single step, so one⌘Z(undo) reverts it if you want the original language back. The translated language is remembered on the deck and shown with a checkmark in the picker.

Translate a single slide: the floating toolbar above every slide has a Translate Slide control. Pick a language to localize just that one slide — handy when you want a single slide in a different language, or to fix one slide without re-running the whole deck. The slide's image is rebuilt so the baked-in text shows the new language. Single-slide translation doesn't change the deck's remembered language, so a deck can intentionally mix languages.

Pick a language when adding a slide: the Add slide dialog has a Slide language picker. It defaults to the deck's current language, so adding a slide to an already-translated deck stays in that language — or pick a different one for just this slide. New slides are authored in English by default; choosing Spanish, French, Italian, or Russian localizes the slide as it's created.

Each slide's visible text is baked into its full-bleed AI image, so translating (whole-deck or single-slide) rebuilds the affected slide images so the on-image text matches the new language. That's why the whole-deck confirmation flags AI-credit usage.

Image Engines: Choosing Between Theo Ultra and Theo Imagine

Every deck carries its own image engine, set from the toolbar pill in the editor and applied to every slide Theo generates. New decks start on Theo Imagine — fast, strong composition, lighter on credits — and you can switch at any time. The choice is per deck, not per account, so a polished client deck and a quick internal one can sit on different engines.

  • Theo Imagine (default): quick and cost-efficient per image, with strong composition and clean in-image text. The right engine for most decks — drafting, long decks where you are generating dozens of slides, and fast iteration on wording and layout.
  • Theo Ultra: the sharpest engine. The strongest typography, the most faithful photography, and the most reliable at complex compositions — a chart, a product shot, and a headline on the same slide. Costs more credits per image and takes a little longer. Switch to it for anything a client, an investor, or an audience will actually see.

A rule of thumb: draft on Theo Imagine, then switch the deck to Theo Ultra and rebuild the slides that matter. Switching engines never touches slides you have already generated — it only changes what the next generation uses.

On the Free plan Theo Ultra appears in the picker with an upgrade link, and every deck generates on Theo Imagine until you upgrade.

Reliability: if your chosen engine has a transient outage, your deck doesn't break. Generation falls back automatically for that one image — ultimately to Theo Creative, the fast engine that backstops the others — and the editor shows a one-time toast so you know the swap happened. Nothing about your deck's setting changes; the next slide goes back to the engine you picked.

Reliable framing: slides are generated with an explicit safe zone so titles and bullets always sit inside the visible 16:9 area, and slide images are normalized to true 16:9 server-side — so the deck you see in the editor matches the file in your export pixel-for-pixel, with no text clipped at the top or bottom edge.

The engine setting also flows through the AI Image Generator modal (when you insert an image directly into a slide) and the Slide Image Studio, so ad-hoc generations match the rest of the deck.

Style-driven Auto-rebuild

When you ask the Global AI Sidebar to rebuild the deck with a specific visual style, the AI locks every slide to that style instead of letting individual slides drift. Phrases like these are recognised automatically:

  • "Rebuild every slide with cinematic style"
  • "Make everything cyberpunk"
  • "Use a watercolour look across the deck"
  • "Switch to anime art on every slide"
  • "Give it a noir film feel"
  • "Vintage 80s VHS aesthetic"

The AI then rebuilds each slide one-by-one, applying the locked style to both copy and imagery so the whole deck stays consistent. Common synonyms — noir, neon, VHS, manga, cel-shaded, watercolour, DSLR, line art, geometric shapes, etc. — also map to the right canonical style.

Style picker apply-to-all: if you change the visual style from the picker and apply it to every slide, the editor offers to also rebuild the imagery so it matches the new style. Confirm to kick off the rebuild, or skip to keep your current images.

Uploading a Logo: The Safe Path

Logos behave differently from every other reference image. Theo never feeds a logo through the image model — the AI would inevitably reinterpret it (slightly wrong proportions, fuzzy typography, off-brand palette). Instead, logos are composited as a deterministic raster overlay on every slide so the asset stays pixel-perfect from slide 1 to slide N.

Canonical flow (recommended): open the slide's Properties panel → Brand section → Upload Logo. This is the byte-for-byte deterministic path — your uploaded image is stored and rendered exactly as you uploaded it. No automatic background removal, no AI synthesis, no edits to the bytes. You can then pick a placement (top-left, top-right, bottom-left, bottom-right) and optionally toggle a rendering style: raw (default — paint the logo verbatim) or chip (wrap the logo in a translucent theme-aware glass backdrop, helpful when an opaque white PNG would otherwise read like a sticker on a dark slide). Both options are opt-in — we never flip Chip on for you.

Tip: if your logo has a tinted background you want to drop, export it as a transparent PNG from your design tool before uploading. Theo will not auto-key the background — it preserves whatever you ship.

Global AI “use this everywhere” also routes through the safe path. When you attach a logo image to the Global AI composer and ask Theo to keep it consistent, the vision classifier flags it as a logo, the server pins it as a logo-mode brand anchor, and the editor automatically composites it via the same overlay used by the Brand Section upload — never through the image model. Trigger phrases Theo recognises:

  • “Use my logo on every slide”
  • “Keep my brand mark consistent”
  • “Add the company logo”
  • “This is our logo — pin it”
  • “Apply the corporate badge”

Per-slide references: logos are refused. The Slide Image Studio modal detects logos in the References tab and refuses to anchor them onto a single slide — instead, an inline banner offers a one-click Set as deck logo CTA that routes the asset through the same deterministic overlay. If you still hit Save & Rebuild while a logo reference is staged, the rebuild payload strips it out so the logo bytes never reach the image model.

Pinning a logo as a brand anchor (via Global AI) is treated identically to a Brand Section upload — the asset becomes theme.logoUrl and every slide's role: "logo" element updates in place. The rendering style (raw vs chip) honours whatever you previously picked in the BrandSection toggle, defaulting to raw. Reload-safe: the apply only fires when the pinned URL differs from the deck's current logo, so re-opening a deck with the same logo is a no-op.

What about Theo Imagine / Theo Creative drift? Image models can still hallucinate fake logos, wordmarks, signatures, monograms, or brand badges in the corners of a slide if asked broadly. Theo's safe-zone directive explicitly forbids the image model from rendering ANY of those, so the reserved corner stays empty for the deterministic overlay to composite on top. If you ever see a duplicated logo on a slide, click Rebuild — the second pass will inherit the strengthened directive and clear the imposter.

Auto-route when you skip the trigger phrases. Natural-language prompts that don't match a logo phrase (e.g. “anywhere you see open charts, use this”) still hit the deterministic path. The Global AI server runs a vision classifier on every attached image and — when it flags the upload as a logo and you didn't explicitly ask to pin it as a brand anchor — emits an appliedAsDeckLogo signal. The editor composites the asset as a corner overlay on every slide and scrubs any “render the {brand} logo prominently” sentences out of the per-slide image prompts so the rebuild loop cannot paint a fake brand mark inside the imagery.

Targeted logo placement: only on slides that match. If you want the logo on a SUBSET of slides instead of every slide, just say so. Theo recognises targeting language and switches from the deck-wide overlay to a per-slide placement. Examples:

  • “Use this logo, but only on slides that mention OpenCharts.”
  • “Add the logo anywhere the slide says ‘Theo’.’”
  • “Pin the company badge for those slides only — not the whole deck.”
  • “Don't use it for all slides — just the ones about pricing.”

When Theo detects targeting language plus a logo upload, the deck-wide overlay path is suppressed and the planner reads every slide's title and body to find the matching slides. Each match gets a per-slide role: "logo" element using the same deterministic raster overlay — never the image model. Unrelated slides stay untouched, and theme.logoUrl is not modified, so the deck-wide overlay does not fire for everything else. You can drag, resize, or remove the per-slide logo on each slide directly.

When no slides match. If the planner can't find any slide that matches your criteria from the deck content, Theo surfaces a warning toast (“No slides matched your targeting — nothing changed”) and leaves every slide untouched. Rephrase the targeting (e.g. name a more specific keyword that you know appears on at least one slide) or open a single slide's AI bar to add the logo there directly.

Refresh existing imagery (one click). When the auto-route fires in a Creator-mode deck that already has rendered slides, Theo offers a confirmation dialog: “Some slides may still show an older AI-painted brand mark inside the image. Rebuild them now?” Confirm to refresh every Creator slide so the underlying scene matches the new overlay; skip to keep your existing imagery and rely on the corner overlay alone.

Logos are the one category of reference image that NEVER touches the AI image model, and your uploaded bytes are preserved as-is. If your logo looks wrong on a slide, re-export from your design tool with a transparent PNG (Theo won't key the background for you) and re-upload via the Brand section.

Brand Anchor: Pin a Reference Image Across the Deck

When you attach an image to Global AI and ask Theo to keep it consistent across the deck, Theo pins the image as a Brand Anchor. From that turn on, every Creator-mode slide regenerates with the actual image bytes feeding the image model — instead of being reimagined from a text description that drifts further each slide.

Three application modes:

  • Subject anchor — the image IS the literal subject (your mascot, character, product hero shot). Every rebuilt slide preserves identity, silhouette, proportions, colors, and core design. Only pose, framing, lighting mood, and surrounding scene adapt to the slide topic. Theo will not silently add wings, capes, accessories, or restyle the subject unless you explicitly ask.
  • Style reference — the image defines the visual style (palette, lighting, rendering technique). Theo renders the subject from your slide directive while matching the reference’s look.
  • Logo — deck-wide logo replacement. Composites the asset as a corner mark across every slide via the existing logo pipeline (the image bytes are not fed to the image model).

Trigger phrases Theo recognises: “use this image”, “use this mascot”, “keep consistent”, “pin as anchor”, “same character”, “don’t change”, and similar variations. Once pinned, the anchor persists on the deck so single-slide rebuilds and “Rebuild all” also use it — you do not have to reattach the image every turn.

Drift-reduction guardrails (subject mode): when a subject anchor is active, Theo runs a multi-engine cascade tuned for identity preservation, in this order:

  1. A purpose-built identity-preserving image edit pass that takes your reference as the literal source frame and only adapts the scene around it.
  2. A high-fidelity multi-image render with the reference (and an isolated background-removed cutout of your subject) sent directly to the image model alongside the slide directive.
  3. A faster multi-image render as the last resort, with the same reference + cutout payload.

Sampling is locked low so identity-defining features (silhouette, palette, accessories) stay stable across rebuilds. The accompanying slide prompt is automatically trimmed to a short composition-only sentence so the verbose theme palette and safe-zone rules can't overwhelm the reference image. Style anchors skip step 1 because they’re matching aesthetics, not a literal subject.

Drift verifier + bounded retry: after each anchored slide renders, Theo compares it back against the pinned reference and scores how faithful the result is. Clean matches ship silently. Visible drift (added accessories, wrong palette, restyled subject) triggers ONE hardened retry that names the drift back to the model so it can correct on the next pass. If the retry still drifts, the slide ships marked anchor weakso you know which slides may want manual attention.

Fallback behaviour: if the reference bytes can’t be loaded or every anchored engine fails, the slide quietly falls back to the standard text-only cascade and the editor surfaces an anchor dropped warning so you can rebuild manually. Logos always use the deterministic logo pipeline, never the anchored image-model path.

Slide Image Studio: Per-slide References, Concept Mode & Engine Picker

The Slide Image Studio (toolbar → Edit Prompt/ Style) is the canonical place to fine-tune one slide's image prompt, attach 1–6 reference images, pick a visual style, and choose which engine renders the rebuild. Unlike the deck-wide Brand Anchor, Studio references apply to just the current slide and the FULL slide text (title + bullets + theme typography) is rendered back into the image.

Three reference modes — each maps to a different image-model contract:

  • Subject — preserve the literal subject pixel-for-pixel. Best for mascots, products, and repeating characters where identity must stay locked across rebuilds.
  • Concept — NEW. Use the reference as inspiration only. The slide preserves the archetype, role, vibe, and count of subjects (e.g. “three people working at a table”, “a confident speaker on stage”) but Theo is free to reinterpret faces, outfits, palette, and rendering medium in the slide's own visual style. Best when you upload photos of real people for inspiration rather than literal replication.
  • Style — match palette, lighting, and rendering technique only. The slide's subject is free to be whatever the prompt asks for. Best for moodboards and aesthetic references.

When you drop a reference, Theo classifies it (People / Mascot / Product / Scene / Logo / Image) and shows a Detected badge on the thumbnail. The mode toggle defaults to the best match — people → Concept, mascots and products → Subject, scenes → Style — and you can override at any time.

Per-rebuild engine picker — Auto, Theo Imagine, or Theo Creative:

  • Auto — inherit the deck-wide image engine. Pick this when you don't need a one-off override.
  • Theo Imagine — the premium engine. Strongest at preserving real people and rendering slide text inside the image. Use it when the reference is a photograph of a person you want to keep recognisable.
  • Theo Creative — the fast Gemini-based engine. Best for mascots, illustrations, and stylistic edits where you want speed + creative reinterpretation over photographic fidelity.

Both modes and engine flow through Save & Rebuild for the current slide only — they don't alter the deck-wide Brand Anchor or imageEnginesettings. Theo respects the slide text on every rebuild so titles, bullets, and theme typography survive instead of getting stripped out for an image-only render.

Choosing references: Upload, From this deck, Mascot & Extract

The Slide Image Studio's Referencessection now opens to a tab strip so you can mix sources without jumping between modals. Every tab feeds the same staged-reference grid underneath, so the first staged item is always the primary anchor for the rebuild — mode toggles, engine picker, and Save & Rebuild work exactly the same regardless of where the reference came from.

  • Upload — the existing drop-zone / file picker / paste handler. Up to six references can be staged in parallel; the classifier tags each one (People / Mascot / Product / Scene / Logo) so the default mode lands somewhere sensible.
  • From this deck — NEW. Shows a tile for every slide in the deck that already has imagery (excluding the slide you're editing). Click a tile to stage it as a style reference — no upload step, no duplicate file in storage. Re-click to remove it again. Use this when you want a new slide to match the palette, lighting, and aesthetic of a slide you've already built.
  • Mascot — NEW. A first-class place to pin a deck-wide brand subject (your mascot, recurring character, hero product). Pinning a mascot stores it on the deck so every NEW slide added afterward inherits it automatically, and a one-click Use on this slide stages it for the current rebuild with the subject-preservation cascade.
  • Extract — NEW. Pull a website's design language (a visual capture plus its palette, type, and a short design brief) and stage it as a style reference for this slide. See Extract From a Website below.

Non-destructive contract for the Mascot tab. This is the most important guarantee — pinning, replacing, or removing a mascot will never silently rewrite slides you've already rendered:

  • Pinning affects only (a) the rebuild you fire next with Use on this slide on, (b) every NEW slide added afterward, and (c) batch rebuilds you explicitly opt into via Apply to all slides… (confirmation dialog).
  • Replacing the mascot swaps the pinned asset in place. Already-rendered slides keep their current images until you rebuild them.
  • Removing the mascot clears the pin. Nothing is rewritten or deleted. Future slides simply stop inheriting it.

When the Mascot is staged for a rebuild, Theo runs the identity-preserving anchored cascade (subject mode) so the mascot stays the literal subject of the slide. Title typography is rendered prominently in-image; bullets and long body copy are intentionally softenedon these rebuilds. Image models reliably garble dense paragraphs when they are also locked to a reference subject (you get duplicated or misspelled words), so Theo will summarise body text to a short supporting tagline or omit it entirely — the mascot + title combination wins every time. If you need the full body text rendered into the slide, run the rebuild without the Mascot staged and re-add the mascot afterward.

Any extra references you stage from Upload or From this deck on the same rebuild still flow through as supplementary cues (composition / palette / mood) without overriding the mascot.

Extract From a Website: Mirror a Site's Design Language

Point Theo at any public website and use that site's design language as the style reference for a slide. Theo captures a visual of the page, reads its palette and typography, and writes a short design brief describing the site's aesthetic — then stages the capture as a style reference so the rebuilt slide mirrors the website's look and feel.

Two ways in:

  • Click the Extract from website globe button on the slide action bar (Creator mode) — it opens the Slide Image Studio directly on the Extract tab.
  • Or open the Studio yourself (Edit Prompt / Style) and switch the References strip to the Extract tab.

Paste a URL (a bare domain like example.com works) and hit Extract. The result card shows the captured visual, the site's palette and fonts, and the design brief. Click Use as design reference to stage it — when you Save & Rebuild, the capture flows through as a style reference and the design brief steers the rebuild alongside it, so the slide picks up the site's palette, type feel, spacing, and overall visual structure.

  • Style, not content: the extraction shapes how the slide looks. Your slide's own title, bullets, and prompt stay the content — nothing is copied off the page.
  • Current slide only: like every Studio reference, it applies to the rebuild you fire next. It never alters the deck-wide Brand Anchor, theme, or other slides. Click the staged card again to remove it.
  • Mix and match: the staged capture sits in the same reference grid as Upload / From this deck / Mascot references, so you can combine it with a mascot or extra style cues on the same rebuild.
The first capture of a site can take up to half a minute; repeat extractions of the same URL are cached and near-instant. If a site can't be captured (blocked, offline, or unreachable), you'll get a clear error and can try another link. This capability is rolling out gradually, so the Extract tab may not be visible on every workspace yet.

AI-Generated Backgrounds

In the slide background picker, click the AI sparkle button and describe the background you want. Examples:

  • "Abstract blue gradient with subtle geometric shapes"
  • "A blurred photograph of a city skyline at sunset"
  • "Minimalist pattern with thin gold lines on dark navy"

The AI returns an image that is automatically set as the slide background. You can regenerate or edit afterward.

Image Styles: 24 Looks for Creator Slides

Every Creator slide is rendered in an image style. Pick one in the Slide Image Studio (or the Design tab’s Image Style card) from 24 looks in four families — each card carries a real thumbnail of the same scene rendered in that style, so you compare styles, not subjects. Apply a style to one slide, or to every slide with a rebuild so the deck reads as one set.

You can also just say it: “make everything pixel art”, “give the deck a Bauhaus look”, “switch to claymation” in the Theo composer locks the whole deck to that style before the rebuild.

Photographic

Camera-real imagery — the right pick when the deck should look shot, not drawn.

Real ImagesReal PeopleCinematicRetro

Illustrated

Hand-made looks, from soft washes to inked panels.

IllustrationsWatercolorAnimeSketchMinimal LineComic BookPaper CollageChalkboard

Stylized

Rendered and sculpted worlds with depth and material.

3D ObjectsIsometricCyberpunkAbstractLow PolyClaymation

Graphic & Retro

Design-history and game-inspired looks with strong, flat colour.

Pixel 8-BitBauhausArt DecoBlueprintRisographVaporwave
Styles shape the imagery; your titles, bullets and layout stay yours. “Pixel perfect” in an instruction is a compliment about sharpness, not a request for Pixel 8-Bit — Theo only switches style when you name one.

Soundtrack Studio (Music & Narration)

Open the Audio panel (music icon in the toolbar) to compose a soundtrack for your deck. The Soundtrack Studio is a simple three-step flow:

  • Pick what you want: Instrumental (an elegant background score by Theo Score) or Song with vocals (an original song by Theo Anthem).
  • Pick a vibe & describe it: 7 curated vibes (Cinematic, Ambient, Upbeat, Lo-fi, Dramatic, Playful, Minimal) plus a Custom vibe where you describe the instruments, mood, and genre in your own words.
  • Lyrics, your way: for songs, write your own lyrics with structure tags ([Verse], [Chorus], [Bridge], [Outro]), tap Write with Theo to have lyrics written from your deck's content, or flip Theo improvises and let the song surprise you.
  • Length: instrumental scores can be 1:00, 1:30, 2:00, or 3:00 — song length follows the lyrics.
  • Library: every generated track lands in your soundtrack library — preview, switch the active track, remove tracks, and set the playback volume.
  • Keeps working in the background: once a track starts rendering you can switch to the Narration tab, start a narration, or close the panel entirely — generation continues, the toolbar music icon pulses while it renders, and a toast tells you when it lands in the library.
  • Seamless playback: the active track fades in when you present, loops gently, ducks under narration during Auto-Present, and fades out when you exit.
  • AI Narration: Theo generates per-slide narration scripts from your content and speaker notes, then synthesizes natural voice audio for each slide. 8 voice personas available — tap the speaker button on any voice card to hear a quick sample of that persona introducing itself before you commit.

See Presenting & Rehearsal → Background Music for playback details.

Instrumental scores cost 500 credits; vocal songs cost 1,000. A lyrics draft from Write with Theo is 100 credits — and you can edit every line before the song is generated.

AI Credits

AI features consume credits from your plan's monthly allotment. The free plan includes 50,000 credits. Credit packs (250K, 500K, 1M) can be purchased separately.

Each operation costs a different amount of credits depending on the model and complexity. Check your usage in Settings → Billing.

Was this article helpful?

Related Articles