Scholarly June 2026 Updates Recap

June was about turning Scholarly into a complete study workspace — more ways to bring material in, more kinds of study material to create from it, and a much more reliable experience on both desktop and phone. We shipped editable library files, Google Drive integration, an illustrated AI Story Book generator, an AI Image Playground, Custom Designs you can reuse across every creation tool, five brand-new document agents, professionally typeset PDFs, and a top-to-bottom mobile redesign — plus dozens of bug fixes across AI Chat, Video Lectures, flashcards, and Deep Research. Here's the full rundown.

---

Major Features

Editable Library Files

Text and data you create with AI — notes, outlines, CSVs, JSON, code, diagrams, and more — are now saved as real files in your library instead of one-off downloads. Each file gets its own page where you can preview it, edit it inline, download it, or restore it if deleted. Open a saved file and chat with it directly: ask questions, request a summary, or have it cleaned up, and the assistant's edits save straight back to the file. Best of all, any saved file can now be turned into flashcards, a quiz, a podcast, a video lecture, slides, a study guide, a worksheet, an infographic, or a mind map — right from the file's own "Create with AI" menu.

How to use it: Ask AI Chat to write something (notes, a CSV, code, a diagram) and it now saves automatically to your library. Open it anytime from Home or search, edit it in place, or hit "Create with AI" to turn it into any other study format.

Google Drive as a Source

You can now create AI content directly from Google Drive — no downloading and re-uploading required. Connect Drive in any create window, pick the exact docs, sheets, slides, PDFs, Word files, PowerPoints, or text files you want, and turn them into flashcards, quizzes, podcasts, AI video lectures, slides, infographics, and more. Drive files also work as sources in AI Chat and Deep Research, and Settings now shows at a glance whether your connection is active or needs reconnecting.

AI Story Book Generator

Turn an idea — or your own PDFs and notes — into an illustrated children's story book. Scholarly writes an age-appropriate story and illustrates every page, keeping the same characters and visual style throughout, then delivers it as a page-by-page book in your library. Choose an art style — Classic Storybook, Soft Watercolor, Bright & Playful, or Dreamy Pastel — and add your own instructions to steer the story. Story Books built from real study material now teach the actual concept at your level, so an exam-prep topic comes back as a genuine explainer, not just a cute story.

AI Image Playground (Beta)

A dedicated space to turn a prompt into study visuals, diagrams, and illustrations. Pick from multiple image models (GPT Image, Grok Imagine, and Google's Nano Banana), choose a style — Educational, General, Photorealistic, Diagram, Infographic, or Artistic — and an aspect ratio, and generate several at once for side-by-side comparison. You can attach up to 4 of your own reference images to remix or edit, and your images keep generating even if you refresh or close the tab.

How to use it: Open the AI Image Playground from Home or the Create menu, describe what you want, pick a style and model(s), and hit generate. Open any result full-screen, download it, or remove it.

Custom Designs

Save your own look as a Custom Design — add style instructions and reference images once, then apply that same look whenever you generate Slides, Story Books, Infographics, Mind Maps, or AI Videos, so everything comes out consistent. Manage your saved designs anytime from Settings, under Scholarly AI.

How to use it: In any create window (Slides, Story Books, Infographics, Mind Maps, AI Videos), open the design picker and choose "Save as Custom Design," or manage existing ones from Settings → Scholarly AI → Custom Designs.

Five New Document AI Agents

We added AI Timeline, AI SOP, AI Flowchart, AI Lesson Plan, and AI Outline — source-grounded PDF agents available from Home, the More menu, chat, upload, and folder create flows. Each one reads your material and builds a structured document tailored to what it's for: a chronological timeline, a step-by-step standard operating procedure, a visual flowchart, a ready-to-teach lesson plan, or a clean topic outline. A new AI Agents page and help guide explain how these agents work.

Professionally Typeset PDF Documents

Study guides, worksheets, lesson plans, outlines, SOPs, and timelines now generate as professionally typeset PDFs — the kind of clean, print-ready handout a professor would give out, with proper headings, callout boxes, tables, and math. Pick a design theme — Professional (default), Academic, Modern, or Playful — right in the create screen to match the look you want.

Quizzes Now Run as Quick Rounds

Quiz sessions now run as short rounds of 10 questions with a recap after each — see your score, redo what you missed, or jump straight into the next round. New decks are ready to quiz on much sooner: the first batch of questions appears while the rest keep generating in the background. Questions also warm you up — starting with easier multiple-choice or true/false before written answers — and the same pacing applies to practice exams, with the Full Exam preset now set to 30 questions in 30 minutes.

Recommended AI Models, Redesigned Picker

The AI model picker across every create tool (flashcards, slides, study guides, AI video lectures, infographics, mind maps, and more) is redesigned: models are grouped into Recommended up top with everything else tucked under Other models, plus a short description under each ("fast and lightweight," "clear, well-paced lectures," etc.). Premium members automatically get the recommended model with no extra setup, and you can now open the picker to browse every available model even on the free plan — picking a non-default model is a paid feature.

Pick Which PDF Pages to Use

You can now choose which PDF pages to include in every AI create flow — flashcards, quizzes, slides, AI video lectures, podcasts, study guides, worksheets, infographics, mind maps, story books, and Study Packs — so you can focus a generation on just the chapter or section you need instead of the whole document.

Deep Research Can Build You Files

Deep Research can now create downloadable files alongside a report — a study guide, a comparison table, a dataset (CSV), a timeline, or a one-page summary. Pick "Add a file" before you run your research. Reports also now include relevant images (diagrams, figures, maps, and photos), inline clickable source pills right in the text, and cleaner formatting with better spacing and headings. Research can also now read your own PDFs, Word docs, PowerPoints, and text files directly as sources — not just images.

Start From Blank

You can now create empty files to fill in yourself — a blank flashcard deck, a blank spreadsheet, and a blank PDF — right from the New menu, which is now split into Blank and Create with AI tabs.

---

AI Video Lectures

Video Lectures had a steady stream of upgrades all month:

• New styles: Editorial (sleek motion-graphics with kinetic typography and animated charts), Retro (80s synthwave/arcade look), and Freestyle (the AI picks the best visual approach for your material) joined Standard, Math, and Kids.
• Faster generation — the lecture builder reads through sources more efficiently and only takes a closer look at pages with diagrams, figures, or equations that actually need it.
• New default fast model (DeepSeek V4 Flash), with existing options still available.
• Video quality control — choose 480p, 720p (default), or 1080p Full HD on paid plans.
• Background music toggle — keep the soft music bed or turn it off for narration over silence.
• Free plan access restored — creating a video simply uses one of your free AI creation credits, same as flashcards, podcasts, and slides.
• Paid-plan logo toggle — subscribers can generate videos without the Scholarly logo in frame.
• Lectures now follow a clearer start-to-finish order (introduce the topic first, keep multi-step processes in sequence) and stay in the language of your source material.

How to use it: Open Create AI Video Lecture, pick a style (Standard, Math, Kids, Editorial, Retro, or Freestyle), set your preferred quality and background music, add your source, and hit Create.

---

Mobile — A Full Pass

June brought a phone-first pass across nearly every surface:

• Phone-native create flows — snap a photo, scan your notes, pick a screenshot, or grab a file, instead of a desktop drag-and-drop box.
• PDFs fit your screen — no more running off the right edge or scrolling sideways.
• Cleaner content headers — just the Scholarly logo, with view switchers (PDF, video, recording) as tidy icon tabs.
• Bigger, easier-to-tap controls everywhere — menus, dialogs, and the plan picker now use the full screen instead of cramped pop-ups.
• Balanced podcast player controls, fixed timestamp wrapping, and a long list of smaller layout fixes across chat, search, settings, Deep Research, studying, and the page editor.
• A refreshed, phone-friendly welcome flow for new sign-ups: pick what you want to make, then snap a photo, choose a file, or paste text, and Scholarly starts building right away.

---

Sharing, Chat, and Everyday Improvements

• Sharing is front and center — saving a video, PDF, flashcard deck, podcast, recording, or research report now opens a quick share panel to send a link in one tap. Shared pages load with a calm branded screen, and Create tools on a shared link now invite visitors to sign up instead of doing nothing.
• Chat shows its work — every step AI Chat takes (including building and reading files) is now visible, so you can follow exactly what it did, and files it creates or reads open in a preview panel next to the conversation instead of a new tab.
• Chat feedback loop — mention a bug or rough edge to the chat assistant, even in passing, and it logs the feedback for our team and confirms it right there.
• Plan with Chat for videos — tap "Plan with Chat" before creating an AI Video Lecture to talk through what to cover, how long, and what to emphasize before it builds.
• A Discord community — join 150,000+ students swapping study tips, linked from Settings and right after finishing a study session, quiz, or exam.
• Marketing pages refreshed — our public website (homepage, pricing, features, and tool pages) got a cleaner, more consistent visual redesign.
• Notification controls — toggle all product and marketing email on or off with a single switch in Settings → Notifications, with one-click unsubscribe on every marketing email.
• The study streak counter was removed from Home and study screens — Scholarly is about making your own material more useful, not daily-streak pressure.

---

How to Use the Highlights

Turn a saved file into study material

1. Open any file in your library (a note, CSV, outline, or diagram you or Chat created).
2. Click "Create with AI" from the file's own menu.
3. Pick flashcards, a quiz, a podcast, slides, or any other format — it builds straight from that file.

Bring in a Google Drive file

1. Open any Create window (flashcards, video lecture, slides, podcast, and more).
2. Connect Google Drive if you haven't already, then pick your file from the picker.
3. Create as usual — no downloading required.

Make an AI Story Book

1. Open AI Story Book from Home or the Create menu.
2. Add a topic, or upload your own PDFs/notes.
3. Pick an art style (Classic Storybook, Soft Watercolor, Bright & Playful, or Dreamy Pastel) and add any instructions.
4. Create — your illustrated book lands in your library, page by page.

Save and reuse a Custom Design

1. In any create window for Slides, Story Books, Infographics, Mind Maps, or AI Videos, open the design picker.
2. Add your style instructions and up to a few reference images, then save it as a Custom Design.
3. Apply it any time you generate one of those formats for a consistent look.

Try a new document agent

1. From Home or the More menu, pick AI Timeline, AI SOP, AI Flowchart, AI Lesson Plan, or AI Outline.
2. Add your source material.
3. Pick a design theme (Professional, Academic, Modern, or Playful) and create — you'll get a clean, typeset PDF.

---

Bug Fixes and Reliability

A long list this month — here are the user-visible highlights:

AI Chat

• Fixed chemistry and math formulas showing raw underscores or codes instead of proper notation (subscripts, fractions, symbols) across chat, quizzes, flashcards, and generated documents.
• Fixed citations sometimes appearing as raw source codes instead of clickable badges, including while an answer is still streaming.
• Fixed AI Chat occasionally getting stuck or failing when asked to build a file — PDF reports, Word documents, PowerPoints, and charts now generate reliably.
• Fixed a rare error that could interrupt a chat after switching AI models mid-conversation.

AI Video Lectures

• Fixed lectures failing to play on iPad and iPhone, and issues where a video's length showed incorrectly while still loading.
• Fixed lectures occasionally coming back in the wrong language, ignoring the requested narration voice, or presenting sections out of order.
• Fixed lectures failing when asked for a much longer video than the source could support — they now produce the best complete lecture the source allows instead of timing out.
• If the fast draft model gets stuck on a longer or more complex source, generation now automatically retries on a more capable model so more lectures finish successfully.

Flashcards and Quizzes

• Fixed large or dense decks sometimes ending up with no practice questions, and decks failing to open during brief server hiccups.
• Fixed vocabulary/translation decks refusing to build from single-language sources, and scanned/image-based PDFs being wrongly rejected.
• Fixed True/False quiz answers coming out almost all "false" — the answer key is now balanced so tests actually check your understanding.
• Flashcards and quizzes now focus on concepts and relationships instead of trivia like page numbers, dates, or where a picture sits.

Uploads and Sources

• Fixed valid PDFs being wrongly rejected, large/poster-format PDFs failing to process, and files becoming unreadable after being moved, renamed, or synced by iCloud/Drive/OneDrive without a clear message.
• Fixed question-bank imports (CSV, Quizlet, Anki, Excel) only importing a fraction of their rows — every question now imports as its own flashcard.
• Clearer, more specific error messages across the board when a source genuinely can't be read, with an inline upgrade path when a limit — not a real error — is the cause.

Deep Research

• Fixed research occasionally finishing with an internal status message instead of a real report, or running for a long time and coming back empty — these cases now retry, and refund your credit if they still can't produce a report.
• Fixed garbled numbers, date ranges, and placeholder links in reports.

Everyday Reliability

• Fixed a crash that could blank out a page (study guides, videos, Home, and others) when a browser was automatically translating Scholarly.
• Fixed a sign-in error on some first-time Google sign-ins, and a rare issue with malformed profile photos after login.
• Chat replies, flashcard generation, and other AI requests no longer fail on a brief network hiccup — they now retry automatically.
• Continued light-mode and dark-mode contrast passes across chat, recordings, podcasts, flashcards, settings, and more, so nothing looks washed out or blends into the page.

Performance

• Faster app responsiveness across Home, chat, folders, and content pages, plus quicker library loading, smoother AI streaming, and lighter, GPU-accelerated animations that respect your system's Reduce Motion setting.

---

Looking Forward

June 2026 was about giving your material more places to live and more ways to become something useful — files you can edit and chat with, Drive as a direct source, a story book generator, an image playground, reusable Custom Designs, five new document agents, and typeset PDFs you'd actually hand out in class. Mobile got a proper pass, and reliability improved across nearly every AI surface.

If you have any questions or feedback, reach out at hello@scholarly.so.