Changelog

Latestproductupdates

Batch Grids, Camera Angles & Text Labels

Editor

Apr 17, 2026

・

v2.19.0 – v2.19.26

This release introduces Grid view for scanning batch × batch generations side-by-side, a full 3D camera angle control in the image sidebar, and Text Labels for annotating your canvas with a Shift + A hotkey.

📦 New

Grid view — chain two batches into a grid of every combination, laid out visually for side-by-side comparison. Pair or Cross any two batches and every cell in the resulting grid is its own generation. Click a cell to open it in Single view and drill in without leaving the grid.
Camera Control — reframe any image from a new perspective with a full 3D angle control. Dial in height, distance, and location of the camera on any image node by going to Tools —> Change camera angles. The generate button shows you how many angles you have queued so you always know an angle is in play.
Text Labels — drop resizable text notes anywhere on the canvas with Shift + A Perfect for labeling sections, leaving comments for collaborators, or organizing complex workflows. Labels resize horizontally for longer notes and pick from a clean size dropdown.
Gemini TTS — Google's Gemini text-to-speech is available on audio nodes. Generate synthesized speech directly on the canvas with voice and language selection built into the node.
Community Tutorials — a Tutorials section in the Community tab with guided project templates. Tutorials open as in-project overlays that walk you through a workflow step-by-step, and completed tutorials land in your FLORA Tutorials folder to revisit later.
Topaz upscaling — the quick upscale action now uses Topaz for sharper, higher-fidelity results. One click on any image node produces a cleanly upscaled version with scale and enhancement tuned for photo-realistic detail.
Per-project canvas position — the canvas now remembers your zoom and pan for every project. Reopen any project and you're exactly where you left off, at the same viewport you had last session.

🧠 Improved

Canvas performance — drag, zoom, bulk-connect, and selection have been rewritten so large projects feel dramatically smoother. Safari canvas zoom in particular is significantly faster.
Error UI — errors now render as a banner directly on the block that failed, with a header indicator and an expandable task bar so you can scan every error across a project at once. Long error messages wrap and scroll cleanly instead of blowing out the block.

🛠 Fixes

Workspace stays mounted across real-time reconnects — no more losing your place on a brief network blip.
Copy as PNG restored in the image node context menu.
Bulk-connect handles no longer orphan after marquee-select moves or ungrouping.
Canvas comments render at the correct constant size on first paint.
Context menu actions freeze during the close animation so you don't accidentally trigger the wrong one.
Audio history items render properly in mobile history views.
Voice Clone surfaces real error messages and guards against oversized uploads.
Audio node modal dismisses with Esc, mic arms before the countdown, and the volume slider has proper spacing.
Parameter sidebar widths no longer run away when a large preview image is loaded.
Fixed an editor crash that could occur with browser auto-translate enabled.
Workspace admins are now treated as editors for project access.
Text blocks no longer flash the prompt when switching between pending generations.

Technique Review Queue & Compaction

Quality of Life

Apr 14, 2026

・

v2.15.88–v2.16.0

A smaller, polish-focused week.

📦 New

Technique Run Counts — see how often a technique has been used, right in the library.
Workspace Member Role Editing — admins can edit member roles directly from the workspace settings.

🧠 Improved

Technique discoverability — richer structured data on public technique pages.

@Mentions & Mobile Generate

Editor

Apr 14, 2026

・

v2.15.33–v2.15.67

This release is a big one. FAUNA's rebuilt narrator onboarding goes live, you can now reference any node directly in your prompts with @, a mobile Generate page ships, and canvas color tags arrive for organizing large workflows.

📦 New

@Mentions in Prompts — type @ anywhere in a prompt box to reference another node on your canvas. A searchable picker appears, and dropping in a chip creates an animated edge from source to target. Chips are click-to-navigate, copy/paste-safe, and resolve correctly across image, text, and video models.
Mobile Generate — a standalone generate flow built for phones, with full-screen model picker, mode switcher, client-side credit checks, and a 4-image default (everything tuned for thumbs).
Canvas Color Tagging — tag blocks with colors, filter the canvas with glow and border highlights, and bulk-download everything sharing a tag.
Model & Parameter Badges in FAUNA chat — see at a glance what's running on each response.
Seedance 2.0 Fast video models available on the model picker.

🧠 Improved

Technique Builder graduated from beta — available to everyone now, with a new onboarding video and smoother exit confirmation
Generation reliability — Gemini rate-limit errors fail fast instead of retrying; fallback routing adapts in real time

Audio Nodes & Technique Publishing

Editor

Apr 6, 2026

・

v2.15.68–v2.15.87

Audio arrives as a full citizen on the canvas. FAUNA chat is roughly 3× faster to first token, and more.

📦 New

Audio Nodes — a dedicated audio node with waveform playback, voice selector, ElevenLabs text-to-speech and speech-to-text, Gemini audio inputs, and automatic routing that picks the right model for music, speech, or sound effects. Perfect for narration, sound design, and podcast-style workflows.
Layer Editor Rewrite — new stencil masking, shapes, text rotation, multi-select resize/rotate, drag-to-adjust numeric inputs (hold and drag any number field), and a proper empty state.
Technique Picker in Add-Node Menu — search techniques with hover-preview thumbnails right from the canvas.
Cmd + **** — Keyboard shortcut to toggle all canvas toolbars to maximize your workspace.
Batch grouping in generation history — variants from the same batch are grouped and highlighted together in the carousel.

🧠 Improved

FAUNA chat — roughly 3× faster to first token
Long FAUNA sessions — now compact automatically, so conversations stay responsive over time

Elements, @ Mentions, & More

Editor

Mar 27, 2026

・

2.15

This release introduces Elements — reusable visual references that live right on your canvas — alongside @ mentions for wiring nodes together, a searchable model menu, full generation history on your dashboard, and major upgrades to batch workflows.

📦 New

Elements — reusable visual references (characters, styles, color palettes) that live as nodes on your canvas. Create them from a right-click or from any selection of existing generations, then pull them into any prompt by typing #. A picker appears, the element's images are automatically fed as input, and the system warns you if you exceed a model's max input count. Learn more about Elements and model compatibility here.
@ Mentions — reference any canvas node directly from a prompt field. Within any prompt field, type @ to open a picker of compatible nodes (connected ones grouped at the top), select one, and a graph edge is created automatically — feeding that node's output into your generation. Works alongside # for elements, giving you two fast ways to wire context into prompts.
Add Node Menu — Model Search — find and add any model instantly. The Add Node menu now includes a search bar across every available model, grouped by provider with submenu expansion. Type to filter, set a quantity for bulk-adding, and models are automatically filtered by compatibility with your current connections.
Dashboard — Generation History — browse every generation you've ever made. A new masonry-grid page on the dashboard shows all past generations across every project with infinite scroll and export options — perfect for revisiting earlier work or pulling assets into new projects.

🧠 Improved

Batch Node — click-to-select sync — clicking a collection item in a batch node now auto-highlights the matching generation in downstream blocks, making it easy to trace results through complex workflows.
Batch Node — downstream workflows — connect a batch node to a full multi-step workflow and run everything end-to-end. Each item flows through all downstream blocks automatically, unlocking true pipeline-style batch processing.

Techniques, Text Layers & Router Node

Editor

Mar 20, 2026

・

2.13

A big one. Techniques brings powerful pre-built workflows you can drop into your canvas or run standalone — think of them as recipes that encode expert-level setups into one click. Text Layers graduates to general availability in the Layer Editor, and the new Router node makes complex canvases dramatically easier to manage.

📦 New

Techniques — use powerful pre-built workflows directly in your canvas or run them independently. Skip the setup and jump straight to results with expert-crafted node chains that encode best practices for common creative tasks.
Text Layers — now available for all users in the Layer Editor. Add and style text layers with full control, opening up typography-driven workflows like posters, social media graphics, and branded content.
Router Node — keep your canvas clean with a simple routing node. Connect one input to your router and fan it out to any number of downstream nodes — hit R to drop one in instantly.

🧠 Improved

Batch Node → Generation Selection — click any collection item in the batch node and it automatically selects the corresponding generation in the downstream node. No more hunting through outputs.
Batch Node Text & CSV Support — use text delimiting or upload a CSV directly to power your batch runs, making it much easier to process structured inputs at scale.
Aspect Ratio Presets — new dropdown in the Layer Editor with your most-used formats baked in. Pick your ratio and keep moving.

Layer Editor, Batch Node, and little big updates

Editor

Mar 13, 2026

・

2.12

Meet the Layer Editor Node, a new compositing toolkit for your canvas. It's also now fully loaded with precision controls, snapping, and keyboard shortcuts that make compositing feel native.

Batch Node lets you process entire queues of assets in one go.

And lots of small upgrades that make your canvas experience feel snappier and more cohesive.

📦 New

Layer Editor v1.0 + v1.1 — a full compositing toolkit for combining multiple images into layered compositions, right on your canvas. Position, resize, reorder, and lock layers with precision. Export the final composition as a single unified image — no need to leave FLORA for basic compositing work. Read about the Layer Editor here.
Batch Node — process multiple assets through a single generation in one run. Feed up to 100 images, videos, or text items into a Batch Node and run them all through the same downstream operation simultaneously. Think bulk upscaling, style transfers, or variations — without clicking "generate" a hundred times. Read more about Batch Node here.

🧠 Improved

Auto Select Mode — the auto-model toggle is now labeled "Auto select mode" for clarity, and defaults to on. FLORA picks the best model for your workflow so you can focus on creating.
Nano Banana 2 as Default Image Model — Google's Gemini 2.5 Flash Image model is now the default for image generation. Fast photo edits, stylized renders, and multi-image fusion at 26 seconds per generation.
Show Credit Balance Toggle — new preference in Account Settings to show or hide your remaining credit balance in the credits indicator. Your workspace, your rules.
Extract Outputs onto Canvas — unpack a node's full generation history into individual nodes. Right-click any node with 2+ completed outputs and select "Extract Outputs onto Canvas." Each output becomes its own node, auto-arranged in a grid next to the source — great for comparing variations or branching your workflow.
Bulk Download — download all generations from a node as a single ZIP. When a node has multiple completed outputs, hit the new download button in the toolbar to grab everything as a neatly named ZIP file. No more downloading one by one.
Expand Text Node Toggle — One click in the toolbar to expand the display dimensions of a text node.
Layer Editor updates
Layer Locking — prevent accidental edits to finalized layers. Toggle lock on any layer via the context menu. Locked layers can't be moved, resized, or deleted, so your careful positioning stays exactly where you put it.
Nudge Layer Positions — pixel-perfect placement with arrow keys. Arrow keys nudge layers by 1px. Hold Shift to nudge by 10px for faster repositioning. Fine-tune your compositions without touching the mouse.
Resize & Center Snapping — layers now snap to frame edges and center points during resize, with snapping enabled by default. Anchor preservation keeps your opposite edge fixed while you drag.
More Aspect Ratios — additional aspect ratio options in the Layer Editor. Corner handles maintain ratio by default; hold Shift to toggle free-form resizing.
Better Frame Border — cleaner, more visible frame border in the Layer Editor for easier spatial orientation while compositing.
Properties Panel Cleanup — reorganized and streamlined with tooltips on every action, so you spend less time hunting for controls.
Dimension Input Cycling — Tab and Shift+Tab now cycle through dimension inputs in the properties panel, committing values as you go. Faster than clicking between fields.

Editing Tools, Usage Insights, & Credit Clarity

Quality of Life

Mar 5, 2026

・

2.11

This release brings a significant upgrade to the editing pipeline — with new inpainting controls, split grid, and smarter canvas interactions — plus a brand new Usage dashboard so you always know where your credits are going.

📦 New

Split into grid — divide any image directly on the canvas into a clean grid of tiles. Available in the Tools dropdown, it lets you break apart compositions, isolate regions for targeted editing, or prep assets for batch workflows — all without leaving your generation pipeline.
Inpaint / Outpaint Variations — generate multiple variation passes on your inpaint and outpaint selections. A new generation stepper lets you dial through variation counts before committing.
Improved Cropping — In-node generation history and crop now work together, so you can move between edits without losing your place.
Usage Dashboard — a full breakdown of your generation history, now live in Settings. See exactly what you've generated, when, and at what cost. Whether you're managing a team budget or just staying on top of personal usage, the dashboard gives you the visibility to plan confidently.
Improved assets — The Assets panel is now the Library. Bring in a full set of references, textures, or source images in a single action and get straight to building. It's larger, includes folders, and allows for bulk upload. Select and upload multiple files at once from the file picker; no more adding assets one at a time.

🧠 Improved

Featured models in the model picker — a curated section surfaces the best new models chosen by the FLORA team
Connected noodle highlighting — selecting any node now illuminates its connected edges, making complex workflows dramatically easier to read and navigate
Refund messaging — when credits are returned to your account, you'll now see clear, timely confirmation of what was refunded and why

Opus 4.6 & Power Tools

Editor

Feb 16, 2026

・

2.11

This release supercharges your creative workflows with Anthropic's most advanced AI model, a brand-new multi-model generation mode for comparing outputs side by side, and a suite of quality-of-life upgrades that give power users fine-grained control over how their workflows run. Plus, video nodes get a major usability boost with inline playback and frame extraction.

📦 New

Opus 4.6 — Anthropic's most intelligent model, now available on FLORA. Brings deeper reasoning, stronger instruction-following, and more nuanced creative output to your text and multimodal workflows — ideal for complex chains that demand precision.

🧠 Improved

Option + Drag to Duplicate — instantly clone any node by holding Option (Alt) and dragging, cutting out extra clicks when you're building or iterating on complex workflows
Lock Node — pin a node's outputs and assets so they won't re-run when you execute the rest of the workflow, perfect for preserving hero assets while experimenting downstream
Copy Prompt — right-click any node to copy its prompt straight to your clipboard for quick reuse or reference
Video Playback Controls — all video nodes now include a scrubber and full playback controls so you can review output without leaving the canvas
Extract Current Frame — grab any frame from a video node as a still image, making it easy to pull reference frames or create thumbnails directly in your workflow
Refund Info on Failed Tasks — the activity queue now shows credit refund details when a task fails, so you always know where your credits stand

🔄 Changed

Styles Sunset (Temporary) — the styles trainer, sidebar access, and style references on models have been removed for now. In practice, state-of-the-art models like Nano Banana Pro already deliver more adherent stylistic results through prompting alone. We're rebuilding styles from the ground up — expect more powerful capabilities like References and Elements for character and style consistency in an upcoming release. In the meantime, you can still upload LoRAs for supported models (Flux Dev, with Flux 2 and Qwen coming soon).

Frame Extraction, Keyboard Shortcuts, & New Models

Editor

Feb 5, 2026

・

2.10

This update makes it faster to work with video and build workflows, plus adds two powerful new AI models. Extract key frames from any video on your canvas, create connected nodes with a single keystroke, and generate with Kling 3.0 and OpenAI o3.

📦 New

Video Frame Extraction — right-click any video to extract its first or last frame as a standalone image. Perfect for pulling start and end frames to use as references, style guides, or inputs for image-to-video generations without leaving your workflow.
Kling 3.0 — Kuaishou's unified multimodal video model with native audio generation and up to 15-second clips. A major leap in cinematic AI video: improved character consistency, multi-shot storyboarding, synchronized dialogue across multiple languages, and significantly cleaner motion in fast-paced scenes. Ideal for narrative-driven video workflows.

🧠 Improved

Keyboard shortcuts for node creation — press T (text), I (image), or V (video) with any node selected to instantly create a new connected node of that type, dramatically speeding up workflow building.

Smarter Defaults & Credit Clarity

Quality of Life

Jan 29, 2026

・

2.9.2

This release is all about reducing friction before you hit Run. You'll now see upfront credit estimates on every generation, start with stronger default models out of the box, and experience fewer timeouts on long-running jobs. Small changes that add up to a smoother, more predictable workflow.

🧠 Improved

Credit Estimates on Generate & Bulk Runs — see exactly what a job will cost before you commit. The Run button now displays a clear credit estimate tooltip upfront, with visual warnings when costs exceed your remaining balance. No more surprise failures mid-batch or guesswork when scaling up expensive workflows.
Stronger Default Models — better results from your very first generation. Image generation now defaults to Flux 2, video and image-to-video default to Seedance 1.5, and inpainting uses Nano Banana Pro.
More Reliable Long-Running Generations — fewer timeouts, steadier progress. Extended jobs that previously timed out due to progress streaming issues should now complete more consistently. We've stabilized progress updates and increased underlying timeouts under the hood, which means fewer interrupted workflows and less need to retry. You'll notice this one when you don't notice it—your heavy jobs just finish.

Usage-Based Pricing

Quality of Life

Jan 27, 2026

・

2.9.1

FLORA's biggest pricing evolution is here. We're eliminating per-seat pricing entirely—your whole team now shares a single credit pool with unlimited members.

📦 New

Credit-Pooled Plans - purchase credits, not seats, and invite your entire team. Share a flexible credit pool across unlimited team members without paying per person. Perfect for growing teams, agencies, and organizations where headcount fluctuates.
Volume Discounts - better rates as your credit needs grow. Scale plan (250K credits) offers significant savings, with additional discounts available on annual billing.

🧠 Improved

All AI models now available on every plan - no more gated access by tier
Downloads, exports, and all features unlocked - previously gated capabilities are now accessible to everyone
Annual billing savings - save up to 17% with yearly commitments across all plans

🔐 Billing Changes

Existing subscribers are automatically grandfathered at current rates indefinitely
No forced migrations or automatic upgrades will occur
Plan modifications are a one-way door—switching to new pricing is permanent

📚 Learn more about usage-based pricing →

Inpaint / Outpaint / Crop

Editor

Jan 25, 2026

・

2.8.5

Edit Where You Generate

This release makes FLORA a complete image creation environment—generate, then perfect without ever leaving the canvas. Split compositions into layers, paint in or out to refine and extend, and crop to focus. Plus, the editor has more indicators to keep you in flow: keyboard shortcuts surface naturally through tooltips, and essential asset context is always visible.

📦 New

Inpainting — replace or remove elements directly on the canvas. Select any region and regenerate just that area. Swap out objects, fix artifacts, or erase distractions without starting over.
Outpainting — extend your images beyond the original frame. Expand the canvas in any direction and let AI fill in the new space. Great for adjusting aspect ratios or giving compositions more room to breathe.
Cropping — narrow in on what matters. Focus your image to the exact frame you need, right where you generated it.
Split into Layers — turn one generation into editable pieces using Qwen's layered edit model. Separate foreground, background, and objects into individual outputs so you can iterate on parts of your composition without regenerating everything. Perfect for refining hero elements while preserving what already works.

To learn more about inpainting and outpainting, visit FLORA Docs.

🤖 New Models

Gemini 3 Flash — Google's latest speed-optimized model. Another option in the speed/quality tradeoff, ideal for rapid iteration when you want faster feedback loops without changing the rest of your workflow.

🧠 Improved

Keyboard shortcuts in tooltips — hover over any toolbar item to discover its shortcut. Power users move faster, and new users learn speed paths naturally while they work.
Info panel visible by default — resolution, cost, model, and other context now shows without hunting. Learn more about the underlying media assets you're working with, at a glance.
Folder labels on dashboard projects — quickly see which folder each project belongs to

Fullscreen History & Sharing Superpowers

Quality of Life

Jan 15, 2026

・

2.7.0

Your generation history gets a major upgrade with a new fullscreen modal featuring project-scoped and all-history views in a beautiful masonry layout. Plus shared projects become more collaborative than ever, and we've added Flux 2 Klein for fast, high-quality generation.

📦 New

Fullscreen History Modal - browse your entire generation history in an expansive, distraction-free view. Launch from the History popover's expand button to access a masonry grid layout with tabs to toggle between current project and all-time history. Perfect for finding that generation from weeks ago without endless scrolling.
Flux 2 Klein - Black Forest Labs' efficient new model optimized for speed without sacrificing quality. Get high-quality results faster with this streamlined Flux variant, ideal for rapid iteration and exploration workflows.
Duplicate Shared Projects - external collaborators can now duplicate shared projects from the context menu. Team members and clients can take shared work and create their own variations without needing full ownership permissions.

🧠 Improved

Shared Projects Sorting - "Shared with me" now sorts by last modified, so your most relevant collaborations surface first
Double-click for fullscreen - double-click any image or video node to open fullscreen preview (also available via F shortcut)
Default Variation Behavior - variations now save "In block history" by default, keeping your canvas cleaner while preserving all outputs
Enhancor v4 Enhancement Modes - strength parameter updated to four intuitive modes: Subtle, Realistic, Pimple, and Freckle (defaults to Realistic)

🛠 Fixes

Veo 3.1 Fast "Auto" aspect ratio now resolves correctly, so image-to-video generations work as expected
Block input thumbnails wrap horizontally again - text selection and clicking work properly without overlapping elements

Model Menu and Full-Screen Generation History

Editor

Jan 6, 2026

・

2.5.3

Navigate our growing model library with ease and revisit your creative journey in a beautiful new full-screen History view. This update brings powerful organization tools to model selection plus a refined way to browse and manage all your generated assets.

📦 New

Model Search - instantly find any model by typing in the model menu. Skip the scrolling and jump straight to the model you need, especially useful as our library continues to expand with new releases.
Pin Models - keep your go-to models at the top of the menu for quick access. Personalize your workflow by pinning the models you use most, reducing clicks and keeping your creative momentum flowing.
Auto Mode - let FLORA recommend the best model for each modality. Turn off manual model selection and get intelligent recommendations based on your content type—perfect for users who want great results without the decision fatigue.
Full-screen Generation History - a completely redesigned view for browsing your generated assets. Access your creative timeline from the History tab in the left sidebar. Review, compare, and revisit previous generations in an immersive full-screen experience.

🧠 Improved

Provider Sorting - models now grouped by provider for easier browsing when multiple models are available from the same source
Double-click Full-screen - quickly expand any node to full-screen view with a double-click for detailed inspection

⚙️ Settings

Prompt Improver Toggle - control whether the background prompt optimizer is applied to your generations. The improver is on by default to enhance prompts sent to models, but you can now disable it in Settings for complete control over your exact prompts.

o3 Deep Research

New Model

Dec 17, 2025

・

2.4.1

OpenAI's research agent for deep competitive and creative analysis lands in FLORA.

🤖 New Models

o3 Deep Research – OpenAI's autonomous research agent for comprehensive multi-source analysis. Searches, interprets, and synthesizes hundreds of online sources—text, images, and PDFs—to produce analyst-grade reports with full citations. The model plans multi-step research trajectories and pivots based on what it discovers, backtracking when needed. In FLORA, use it to kickstart creative briefs with competitive landscape analysis, gather reference material across dozens of sources, or research trends before building your workflow. What used to take hours of tab-juggling now feeds directly into your canvas.

🧠 Why Deep Research in FLORA?

Creative work increasingly starts with research—understanding what exists, what competitors are doing, and what references to draw from. o3 Deep Research lets you front-load that discovery phase directly in your workflow. Ask it to analyze visual trends in a category, compile brand positioning across competitors, or gather technical specifications for a product shoot. The output flows naturally into your creative process, giving you a research foundation without leaving FLORA.

GPT Image 1.5, Flux 2 Max, & 2FA

New Model

Dec 16, 2025

・

2.4.0

Three major image and video models land this release, each pushing the boundaries of what's possible in a single generation. Plus, optional two-factor authentication for teams that need tighter account security.

🤖 New Models

GPT Image 1.5 – OpenAI's flagship image model with dramatically improved editing precision. Up to 4x faster than its predecessor with best-in-class instruction following. Excels at targeted edits that change only what you ask for—adjusting lighting, swapping outfits, or refining expressions—while preserving everything else. Significantly improved text rendering handles denser, smaller type for mockups and marketing materials.
Wan 2.6 – Alibaba's cinematic video model with multi-shot storytelling and synchronized audio. Generate up to 15 seconds of 1080p video with lip-synced dialogue, sound effects, and music in a single pass. Reference-driven generation keeps characters, voices, and styles consistent across shots—ideal for ads, product videos, and narrative content without post-production.
Flux 2 Max – Black Forest Labs' top-tier image model for maximum quality and control. The most capable FLUX variant yet, with the strongest prompt adherence and highest editing consistency in the lineup. Reference up to 10 images to maintain character identity, product details, and brand elements across batches. Outputs at up to 4MP with photorealistic lighting and physics—perfect for product imagery, marketing assets, and any work where consistency matters.

📦 New

Two-factor authentication – optional 2FA for stronger account security. Enable in your account settings for an extra layer of protection on your workspace.
Replace image in toolbar – swap assets directly from the block toolbar. Faster iteration when you need to update a source image without opening menus or rewiring connections.

Credit Limits

Quality of Life

Dec 15, 2025

・

2.3.6

Give your team access to AI generation without giving away the budget. Set per-user credit limits directly from Admin, with full visibility into who's using what — perfect for agencies managing client access or teams scaling up without surprises.

📦 New

Per-user credit limits – Set spending caps for any user from Settings → Workspace → People & Credits. Choose one-time limits for project-based work or recurring limits that reset monthly. Your team gets creative freedom; you keep budget control.
Usage visibility – See credit consumption at the user level, so you always know where resources are going before the bill arrives. No more end-of-month surprises.

In-Node Generation History

Editor

Dec 10, 2025

・

2.3.4

Every node now remembers its history. Browse, compare, and reuse previous outputs without cluttering your canvas or losing that perfect variation from three iterations ago.

📦 New

In-node generation history – Every node stores up to 100 past generations. Scroll through them in a carousel, navigate with keyboard shortcuts or trackpad gestures, and duplicate or export any result from the new menu.
Default variation behavior – Choose how variations work for you. Head to Project Settings → Preferences → Variation Behavior to set whether new generations appear on canvas or stay in-node by default. Switch anytime based on how you like to work.
Updated full-screen view — Hit 'F' on any block to see an expanded view that includes asset metadata, like file size, resolution, credit cost, and more. This view also has a grid view so you can view all assets from one node.

Next-Gen Text Models

Editor

Dec 4, 2025

・

2.2.6

A major model refresh brings the latest from ByteDance, Kuaishou, OpenAI, and Anthropic.

🤖 New Models

Seedream 4.5 – ByteDance's commercial-grade image model with exceptional text rendering. Sharp, legible typography for posters, logos, and marketing layouts. Generates multiple consistent images from a single prompt while maintaining character and environment stability across the batch—up to 4K output.
Kling 2.6 Pro – Kuaishou's latest video generation upgrade. Improved motion quality and scene coherence over 2.5, with better handling of complex camera movements and longer output durations.
GPT-5.1 – OpenAI's newest reasoning model for prompt intelligence. Enhanced instruction-following and creative ideation, powering smarter prompt suggestions and workflow assistance throughout FLORA.
Claude Opus 4.5 – Anthropic's most intelligent model with exceptional depth and nuance. Extended thinking capabilities for complex creative briefs and multi-step reasoning. When you need careful, thoughtful output, Opus 4.5 delivers.
Claude Sonnet 4.5 – Anthropic's balanced model for speed and intelligence. Fast enough for real-time assistance, smart enough to meaningfully improve your prompts and workflows. The engine behind much of FLORA's responsive intelligence.

Kling O1 & Qwen Edit Angles

Editor

Dec 2, 2025

・

2.2.2

Two powerful new models expand your creative toolkit: Kuaishou's latest Kling O1 family brings improved video generation across multiple modalities, while Qwen Edit Angles gives you precise camera control for image editing. This release also delivers a major collaboration upgrade with proper private sharing and refreshed navigation throughout.

🤖 New Models

Kling O1 – Kuaishou's latest model family with expanded capabilities across modalities. Includes text-to-video, image-to-image, image-to-video, and edit variants. Improved motion coherence and prompt understanding over the 2.5 generation, with stronger performance on complex scenes and longer outputs.
Qwen Edit Angles – image-to-image model for precise camera viewpoint control. Adjust perspective, angle, and viewpoint on existing images with more control than traditional editing tools. Great for product shots, architectural visualization, and creative reframing.

📦 New

Private project sharing – share private projects via editable links without making them public. Invite specific collaborators, including external users and people from other workspaces, while keeping your work hidden from the broader workspace. True document-style privacy with selective access.
Sharing for free users – all users can now share projects via links. Basic sharing is now available to everyone.
Drag-to-replace on blocks – drop a local file onto an existing image or video block to swap its asset. Faster iteration without rewiring connections—just drag your new file directly onto the block you want to update.

🧠 Improved

Project & workspace breadcrumbs in the header for clearer navigation context
Pagination restored across all workspace views with proper load-more behavior
Tool popovers (Assets, History, Groups, Styles) now open on click with tooltips, replacing the hover behavior
Delete key now mirrors Backspace for block deletion—no more accidental deletions while typing in prompt fields
Geist Sans is now the global typeface for a cleaner, more modern feel
Workspace switching on invite accept – if you're already a member, accepting an invite now switches you to the right workspace automatically

🛠 Fixes

Folder and workspace views now correctly enforce membership permissions
Fixed crash when copying groups with regex-special characters in their names
Improved in-app notifications

Project Sharing & Open Collaboration

Editor

Nov 26, 2025

・

2.1.8

We're excited to launch a major update to how you share and collaborate in FLORA. You can now share your projects with anyone—inside or outside your workspace—with flexible permissions that put you in control of how others view, edit, and remix your workflows.

📦 New

Project-Level Sharing - share any FLORA project with granular permission controls. Choose whether your project is restricted to only specific people you invite, or whether anyone with a link can view or edit your project. Perfect for sharing workflows on social media, internal Slack channels, or with clients—let others clone and remix your work with specific permissions.
Cross-Workspace Collaboration - invite collaborators without adding paid seats. Now, you can share projects with users outside your workspace and let them use their own credits for generations. No more seat management headaches—just send a link and start creating together. Just share a link or invite an editor to collaborate, and watch the shared work unfold live with real-time collaboration.

We're rolling out additional billing updates over the next few weeks to make collaboration even smoother. In the meantime, we'd love to see what you share—tag us when you publish your workflows!

Keyboard Shortcuts

Quality of Life

Nov 25, 2025

・

v2.1.5

Power users rejoice: comprehensive keyboard shortcuts now cover nearly every basic canvas action, with a built-in reference guide to help you work faster.

📦 New

Keyboard shortcuts – full keybind support for common canvas actions. Add image, video, and text blocks, group/ungroup selections, select all, copy, duplicate, delete, rename, fullscreen preview, upload, and remove background—all from the keyboard. Check the new "Keyboard Shortcuts" entry in the Help menu for the complete reference.

🧠 Improved

Generation history queries faster for project-level sharing with a new database index
VEO 3 image-to-video now preserves your chosen aspect ratio through to the final request

🛠 Fixes

Multiplayer undo/redo temporarily disabled with a clear toast explanation
"Learn about Styles" link in the toolbar now points to the correct documentation

Flux 2 Arrives

New Model

Nov 25, 2025

・

2.1.4

Black Forest Labs' next-generation image model lands on FLORA with full multi-modal support, alongside smarter image loading that keeps your canvas responsive even with high-resolution assets.

📦 New

Flux 2 – Black Forest Labs' latest image generation model across all input modes. Available in both Dev and Pro tiers with support for text-to-image, image-to-image, and image+sketch-to-image workflows. Delivers improved prompt adherence and visual quality over the original Flux lineup.
Flux 2 Dev — now supports up to 3 inputs for more complex compositional workflows. Combine multiple reference images to guide your generations with greater creative control.

🧠 Improved

Smart image loading (LOD) – images now load at the appropriate resolution for your current zoom level. No more waiting for full-resolution assets when you're zoomed out. The canvas stays snappy by requesting only the detail you need, with the next resolution tier preloading in the background as you zoom in.
Image blocks now carry width/height metadata, preventing unnecessary high-resolution requests for smaller source images

LTXV 2 & Zoom Enhancements

New Model

Nov 25, 2025

・

2.1.3

The latest video model from Lightricks lands alongside better zoom controls and under-the-hood fixes for model selection and integrations.

📦 New

LTXV 2 – Lightricks' updated video generation model with improved quality. Enhanced output fidelity and better prompt adherence for more predictable, professional results.

🧠 Improved

Deeper zoom now available on blocks – inspect fine details without losing context
Variations button removed from toolbar for a cleaner interface

🛠 Fixes

Blocked models now correctly enforced in backend – no more unexpected model availability issues

Nano Banana Pro

New Model

Nov 20, 2025

・

2.1.1

Next-Gen Image Generation

Experience a massive leap forward in AI image generation with Nano Banana Pro. Built on the powerful Gemini 3 architecture, this model delivers unprecedented precision in text rendering, deeper contextual understanding, and exceptional control over complex visual layouts.

📦 New

Nano Banana Pro - Google's next-generation image model with state-of-the-art text integration Finally achieve crisp, accurate typography within your generated images. Combined with advanced world knowledge for better contextual understanding and exceptional performance on detailed infographics and complex layouts, this model excels at prompts that require precision and nuance.

Canvas at scale

Quality of Life

Nov 15, 2025

・

2.0.1

This release supercharges FLORA for power users working with complex, large-scale projects. Experience dramatically faster performance on canvases with hundreds of nodes, handle massive 1000+ node projects, and take control of your workspace with new bulk operations and asset management tools.

📦 New

Bulk Add Blocks - rapidly populate your canvas with multiple blocks at once. Perfect for setting up complex workflows or duplicating patterns across your project - spend less time on repetitive setup and more time on creative execution.
Delete Saved Assets - full control over your workspace storage and generation history. Clean up your workspace by removing unwanted saved assets and past generations, keeping your projects organized and focused on what matters.

🧠 Improved

Canvas performance optimized for hundreds of nodes - smooth, responsive editing even on the most complex projects with multiple concurrent generations running.
Large canvas support enhanced for projects exceeding 1200 nodes - seamlessly work on enterprise-scale canvases without performance degradation or stability issues.

Workspace Polish & Navigation

Quality of Life

Nov 10, 2025

・

1.33.0

This release focuses on refining your workspace experience with smarter model management, cleaner navigation, and more intuitive credit notifications. We've smoothed out rough edges in the interface and made view-only mode more professional for client presentations.

🧠 Improved

Blocked Models interface for admins now shows the model trainer instead of provider - making it clearer who created each model, with updated search functionality and stronger backend enforcement for workspace compliance and brand safety policies
Canvas navigation eliminates context menu flashing when panning with spacebar - enjoy smoother, distraction-free movement around large canvases during creative work
Credit notifications are now smarter with autopay - when automatic payments are enabled, you won't see disruptive "out of credits" alerts since your account will seamlessly replenish
View-only mode hides editing toolbars for cleaner presentations - share your work with clients and stakeholders without distracting interface elements, perfect for professional reviews and approvals

Model Access Control for Teams

Editor

Nov 1, 2025

・

1.32.0

Workspace admins can now control exactly which AI models their team can access. This highly requested enterprise feature lets you customize your workspace to match your organization's needs, budget, and creative standards—restricted models are completely hidden from member workflows.

📦 New

Model Access Control - workspace admins can now enable or disable specific models for their team. Perfect for agencies managing client work standards, studios controlling costs, or enterprises with compliance requirements. Restricted models are completely hidden from the Editor, keeping your team focused on approved tools. To access: admins can go to Settings > Workspace Settings and toggle models.
Minimax 2.3 - enhanced text generation with improved coherence and speed. Delivers sharper, more contextually aware outputs while processing faster than previous versions, ideal for rapid ideation and content workflows.

🧠 Improved

VEO 3 image-to-video now supports flexible aspect ratio parameters - transform your stills into motion with precise control over output dimensions
Wan 2.5 duration control - set custom video lengths to match your exact project needs
Generation queue handles concurrent tasks more intelligently, prioritizing based on resource availability
Improved processing of credit refunds for failed generations

Veo 3.1

New Model

Oct 15, 2025

・

1.30.0

VEO 3.1 delivers production-quality video with synchronized audio built-in, perfect for professional storytelling and cinematic workflows on FLORA.

📦 New

VEO 3.1 - Enhanced realism with true-to-life textures, stronger prompt adherence, and native audio generation including natural conversations and synchronized sound effects

🛠 Fixes

Video preprocessing pipeline optimized for smoother uploads

Seedance Pro Last Frame & Quality Fixes

Quality of Life

Oct 9, 2025

・

1.29.0

Control your video endings with Seedance Pro's last frame feature, upscale to even higher resolutions with Magnific, and enjoy more reliable generations overall.

📦 New

Seedance 1.0 Pro last frame - precise control over video endings
Version number now links to updates page

🧠 Improved

Magnific maximum size increased for higher resolution outputs
Pre-generation timeout issues resolved

🛠 Fixes

Email change functionality fixes
Generation stability improvements

Sora 2 Now Live

New Model

Oct 7, 2025

・

1.28.3

📦 New

Sora 2 - The latest video-gen model from OpenAI. Create high-quality user-generated content with voiceover and music directly from a single image and prompt.

Some notes on SORA 2 - the API enforces several content restrictions:

Only content suitable for audiences under 18 (a setting to bypass this restriction will be available in the future).
Copyrighted characters and copyrighted music will be rejected.
Real people—including public figures—cannot be generated.
Input images with faces of humans are currently rejected.

Make sure prompts, reference images, and transcripts respect these rules to avoid failed generations.

Nano Banana Aspect Ratios

Quality of Life

Oct 2, 2025

・

1.28.0

📦 New

Aspect ratio support: Nano Banana (Gemini-2.5 Flash Image) now supports defined aspect ratios! Select from a range of aspect ratios in the parameters toolbar when editing images.
Content out of view toast - never lose track of nodes on large canvases - we now help you zoom back to your content when you scroll away
Better signup flow - OTP verification replacing magic links for faster, more reliable signup

Replace Block and New Models

Editor

Sep 24, 2025

・

1.27.0

Improved workflow editing with Replace Block and some brand-new AI models.

Replace Block Transform your workflows on the fly! The new Replace Block feature lets you swap media in existing workflows without rebuilding from scratch - just right click on any image or video block and drop in your new content. We'd recommend trying this feature on any existing group — use pre-built groups to run workflows with new assets.

New models

Explore creative possibilities with Kling 2.5, WAN 2.2 Replace and Move, and Lucky Edit Pro for professional video editing.

📦 New

Replace Block - seamlessly drop new media into existing workflows
Kling 2.5 - latest version with enhanced generation quality
WAN 2.2 Replace and Move - advanced object manipulation capabilities
Lucky Edit Pro (v2v) - professional-grade video-to-video editing

🧠 Improved

Seedream preprocessing for input images
Updated placeholder text for mixed-to-video operations
Clearer communication about inactive features for Free users

🛠 Fixes

Onboarding redirect client-side exception resolved
Various stability improvements

New landing page and onboarding updates

Quality of Life

Sep 18, 2025

・

1.26.0

New user welcome experience and modernized landing page.

📦 New

Welcome onboarding screen - friendly introduction for first-time users
Framer-powered landing page - completely redesigned with modern animations and interactions

🧠 Improved

ImageKit transformation URL function refactored for better performance and reliability
Smoother image processing and delivery across the platform

Breakout Lists & Style Control

Quality of Life

Sep 10, 2025

・

v1.24.0

📦 New

Split list into blocks - Automatically split any list in a text block into a series of blocks! To activate, go to any text block that has a list in its output. You'll see a "split list into blocks" button in the toolbar on hover. This will spawn out a series of nodes — each one has a list item within. Great for text to prompt to image / video workflows.
Ideogram Styles - Access curated style presents from Ideogram on any text-to-image prompt with the Ideogram model

Infrastructure Optimization

Quality of Life

Sep 4, 2025

・

v1.23.0

Backend infrastructure improvements for faster generation and better stability.

🧠 Improved

Expanded queue capacity for Nano Banana and Flux Kontext - reduced wait times
Liveblocks upgrade for enhanced real-time collaboration
Nano Banana image queue optimization for smoother processing

GPT-5 and Model Expansion

Quality of Life

Aug 27, 2025

・

v1.22.0

Major AI model upgrades including GPT-5 and new creative tools.

📦 New

GPT-5 as default text model - experience the latest in language AI
Magnific API integration for advanced image enhancement
Seedance 1.0 for dynamic generation workflows
Higgsfield T2I model for text-to-image creation
Recraft SVG tool for vector graphics generation

🧠 Improved

Aleph prompt length validation - smart handling of long prompts
Human-readable Runway error codes for clearer troubleshooting
Enhanced model selection UI with scrollable interface

🛠 Fixes

Onboarding bugs causing users to get stuck on stages resolved
Liveblocks authentication logging for better debugging
Various stability improvements

Moonvalley Marey

New Model

Aug 21, 2025

Marey is Moonvalley’s foundational AI model, built to meet the standards of world-class cinematography.

WAN 2.2

New Model

Aug 19, 2025

Wan 2.2 is a cutting-edge open-source AI video and image generation model developed by Alibaba, featuring major advances in cinematic aesthetics, motion generation, and usability for both text-to-video and image-to-video workflows.

New Models

New Model

Aug 18, 2025

・

v1.20.0

Expanding our creative capabilities with three new models.

📦 New

WAN 2.2 - enhanced generation capabilities
Seedream - for imaginative visual creation
Moonvalley Marey - model integration for specialized video workflows

Character Generation

Quality of Life

Aug 7, 2025

・

v1.19.0

📦 New

Ideogram Character support for consistent character generation

🧠 Improved

Changed default GPT input fidelity for faster initial generations and cost optimization

Canvas Stability Fix

Quality of Life

Jul 30, 2025

・

v1.18.0

🛠 Fixes

Lost progress on projects resolved - Liveblocks synchronization fixed
Canvas now properly saves and persists all work

Onboarding Update

Quality of Life

Jul 27, 2025

・

v1.17.0

🛠 Fixes

Onboarding flow corrected - new accounts without projects now properly see onboarding instead of being sent directly to dashboard
Small visual refinements throughout the interface

Collaboration Improvements

Editor

Jul 16, 2025

・

v1.16.0

📦 New

Upload option in node menu - directly add files to your canvas workflow
Auto-updating version display - always know you're on the latest build

🧠 Improved

Complete typography overhaul for better readability and visual hierarchy
Generate button is now less sticky for smoother scrolling
Enhanced onboarding with strategic interaction blocking
Enterprise credit system now functioning perfectly

🛠 Fixes

Groups dropdown selection bug resolved
Enterprise credit awarding issues completely fixed
Various UI interaction improvements

History & Regional Access

Editor

Jul 9, 2025

・

v1.15.0

Enhanced generation history, regional controls, and canvas organization improvements.

📦 New

Multi-select projects on dashboard for bulk operations
Alpha transparency checker pattern for transparent images
Block resizing for flexible canvas organization
Luma Modify Video integration for advanced video editing
Regional access controls for India and Indonesia

🧠 Improved

Generation history fully restored - properly hydrates all past work
Significant performance improvements across the platform
Expanded Dub integration features

🔐 API Changes

Regional paywall implementation for specific markets
Enhanced Stripe email handling

🛠 Fixes

Sanitized email issues in Stripe integration resolved
DOCTYPE rendering errors eliminated
Various stability improvements

Smart Defaults & Teams

Editor

Jul 3, 2025

・

v1.14.0

Introducing intelligent workflow defaults and team collaboration features.

📦 New

Global settings for default models - set your preferences once, use everywhere
Groups for team organization and permission management
Onboarding v2 - streamlined experience for new users
Larger language models support for advanced text generation
Dub partners integration for enhanced link management

🧠 Improved

Flux Dev T2I on FAL - faster and more cost-effective
Enhanced Flux Kontext prompt improver
Bulk generation blocks now reset to 1 for easier re-running
Clearer warnings about editor invitation costs
Empty prompt validation - prevents accidental blank generations

🛠 Fixes

Copy/paste now properly retrieves parent nodes
Multiple styling issues resolved
Numerous backend stability improvements

Metadata & User Feedback

Editor

Jun 25, 2025

・

v1.13.0

Enhanced file handling with metadata support and improved user feedback channels.

📦 New

Suggest a Feature button - share your ideas directly with our team
PNG metadata embedding - generation settings preserved in exported files
ESLint implementation (phase 1) for improved code quality

🧠 Improved

Run All command now intelligently excludes utility nodes
Performance boost - toolbar only appears when nodes are actively focused
Updated privacy policy documentation link

🛠 Fixes

Multiple dashboard display bugs resolved
Toolbar persistence issues fixed
Various performance optimizations

Video Evolution

Quality of Life

Jun 18, 2025

・

v1.12.0

Enterprise authentication and major video processing improvements with audio support.

📦 New

Enterprise SSO/SAML authentication for seamless corporate integration
Hailuo 2 model for next-generation AI capabilities
Audio-preserved video upscaling - enhanced videos now retain original sound

🧠 Improved

Textarea autosize performance significantly optimized
Video processing pipeline upgraded to maintain audio tracks

🔐 API Changes

Enhanced security protocols for corporate deployments

VEO3 Now Live

New Model

Jun 5, 2025

・

v1.10.1

📦 New

VEO3 integration - Google's latest video generation model

🧠 Improved

Prettier code formatting implemented across the codebase
Cleaner, more maintainable code structure

🛠 Fixes

Share button functionality restored
Various minor bug fixes and stability improvements

Kling 2.1 & Admin Tools

New Model

May 30, 2025

・

v1.9.6

New AI model integrations and enhanced administrative capabilities (catching up on missed updates!).

📦 New

Kling 2.1 model integration for advanced generation capabilities
Kontext for intelligent prompt enhancement
Community admin dashboard for better content moderation
Default GPT image placeholder implementation

🧠 Improved

Pricing page redesign with comparison table
Dashboard visual enhancements for better usability
Pricing structure visualization improvements

🛠 Fixes

Green background rendering issue resolved
Various visual bugs on dashboard and pricing pages

Performance Boost

Quality of Life

May 22, 2025

・

v1.8.0

Enhanced moderation capabilities and generation speed improvements with Runway integrations.

📦 New

Flux Pro moderation for enterprise-grade content filtering
Runway References for style-based generation workflows
Updated manifesto and refined landing page

🧠 Improved

Faster LoRA generations - significant speed boost for custom styles
Runway Gen3 aspect ratio handling corrected
Drop cursor positioning improved with right-side icon alignment

🔐 API Changes

Generation stopping disabled from active tasks menu for workflow consistency

🛠 Fixes

Various landing page display refinements
Aspect ratio issues with Runway Gen3 resolved
Cursor positioning improvements in the canvas

Video Upscaler Hotfix

Quality of Life

May 16, 2025

・

v1.7.1

Quick fixes for video processing and LoRA functionality.

🛠 Fixes

Video upscaler critical hotfix deployed
Noodle connection issues resolved
Minor LoRA upload/download improvements for stability

LoRA Updates

Quality of Life

May 14, 2025

・

v1.7.0

Custom style support with LoRA upload/download capabilities and LLM infrastructure upgrade.

📦 New

Upload/Download custom LoRAs for personalized style transfer
Geist Mono as default monospace font for better code readability
Non-temporary URLs for reliable endpoint access

🧠 Improved

Gemini migrated to Vertex for enhanced LLM service reliability
Better error page design with clearer messaging
Text-to-video aspect ratio handling with styles

🛠 Fixes

Video and text blocks now properly restricted in Create Style mode
Video variations removed for improved stability
Various endpoint URL persistence issues resolved

Performance Sprint

Quality of Life

May 10, 2025

・

v1.6.2

Performance improvements and UI refinements across the platform.

🧠 Improved

Input length checker to prevent oversized prompts
40% faster trackpad scrolling for smoother navigation
Liveblocks optimistic loading now functioning as intended
Landing page visual updates and fixes

🛠 Fixes

Hover toolbar disabled for cleaner interface
View-only mode no longer shows hover toolbar options
Broken community thumbnails restored
Various landing page display issues resolved

Streamlined Generation

Quality of Life

May 8, 2025

・

v1.6.1

📦 New

Automatic tax collection enabled for all transactions

🧠 Improved

Generation cancellation removed - commits to completion for better reliability

🛠 Fixes

Blinking caret issue in text inputs resolved

Account Management

Quality of Life

May 7, 2025

・

v1.6.0

Enhanced account handling and generation control improvements.

🧠 Improved

Sign-in page visual refresh
Faster "Copy Share Link" functionality
Better deleted account handling
Modal polling with backend error logging
Stop generation on block delete - automatic cleanup
Expanded username formats (beyond alphanumeric)

🛠 Fixes

User subscription edge cases resolved
Project title truncation issues
Input size guardrails implemented
Community overflow display issues

Saturday Performance Special

Quality of Life

May 3, 2025

・

v1.5.0

Weekend performance boost with significant speed improvements.

🧠 Improved

Major performance boost - increased viewport node mounting threshold
In-editor notification polish

Generation Quality Upgrade

Quality of Life

Apr 30, 2025

・

v1.4.0

Major improvements to generation quality and multiplayer experience, with significant backend optimizations.

📦 New

Editor notifications for real-time updates and credit returns
Posthog analytics for download tracking from nodes
Backend migration for improved performance and reliability

🧠 Improved

Dramatically reduced bad generations on OpenAI requests - our biggest quality improvement yet
Gemini model enhancements for better outputs
Visual redesign of top navigation bars
Endpoint dropdown with clearer visual hierarchy
Block titles now auto-populate on connected nodes