Kolbo.AI is the most comprehensive AI creative platform that consolidates 100+ leading AI models and tools into one unified workspace. Create images, videos, music, audio, 3D models, and manage complex workflows without switching between multiple platforms. Everything from ChatGPT and Claude to Midjourney, Kling, Suno, and Eleven Labs—all working together with shared context and memories.

Does Kolbo.AI have a desktop app and Adobe integrations?

Yes! Kolbo.AI offers a desktop app plus plugins for Adobe Premiere Pro and After Effects. This means you can drag and drop AI-generated content directly onto your video editing timeline, streamlining your creative workflow. This is a unique feature that competitors don't offer.

What happens to my unused credits?

Your credits roll over to the next month as long as you don't cancel your plan! This is a unique Kolbo.AI feature—unused credits accumulate, so you never lose what you've paid for. No other competitor offers this benefit.

What AI models do you have in the platform?

We offer the biggest range of AI models in the industry—100+ models carefully curated for production work. This includes: ChatGPT (GPT-4), Claude (Sonnet 3.5), Gemini Pro, Midjourney, Flux, Seedance, Ideogram, Kling, Hailuo 2.3, Sora 2, Veo 3, Suno, Eleven Labs, DeepDub, and many more. We monitor and add new models daily, ensuring you always have access to the best and most advanced AI technology. See our complete AI Models showcase for the full list with visual icons and categories.

What can I do with Kolbo.AI?

Create professional content across all media types: (1) Images: Text-to-image with 20+ models, train custom models, use Angels for multi-angle shots, Canvas for advanced editing, Creative Director for complete creative sets. (2) Videos: Image-to-video, text-to-video, Lipsync, Motion Transfer, Character Replace, First Last Frame, Elements, video editing (upscale 4K, reframe, remove bg). (3) Audio: Music with Suno, text-to-speech with voice cloning (Eleven Labs, DeepDub), transcription, subtitles. (4) 3D Models: Generate 3D assets from text or images. (5) Chat: Talk with any LLM, create custom agents, use the Artifacts window for code/docs, compare multiple models. All organized in project-based workflows with shared context.

Why choose Kolbo.AI over competitors like Higgsfield, Krea, Freepik, or others?

Kolbo.AI is the only platform that combines: (1) 100+ AI models in one workspace with shared context. (2) Desktop app + Adobe plugins for professional workflows. (3) Credits that roll over monthly. (4) Complete creative toolkit: images, videos, 3D models, audio, music, and chat. (5) Unique features: Angels, Creative Director, Training Lab, Multi-Model Comparison, Unlimited Concurrent Generations. (6) Deep collaboration with full session/project sharing. (7) Professional tools: video editing, custom model training, Artifacts window. Competitors like Midjourney, Runway, or ChatGPT focus on single modalities. Kolbo.AI unifies everything with industry-leading project management, so you never have to piece together different apps or lose context.

What makes Kolbo.AI unique?

Kolbo.AI stands out because all AI models work with the same shared context—your conversations, memories, and projects. Every tool, from text to image to video to 3D, builds on your current work. Our project management is the most comprehensive you'll find: each project contains organized sessions for every tool. Plus, unique features like the Desktop App with Adobe plugins, Credits Roll Over, Multi-Model Comparison, Angels (multi-angle generation), Creative Director, Training Lab, and Unlimited Concurrent Generations set us apart from competitors.

Do I need any setup or API keys?

No setup and no API keys required—Kolbo.AI is ready to use as soon as you sign up. We handle all the technical complexity so you can focus on creating.

What is the Angels feature?

Angels is a unique Kolbo.AI feature that generates multiple angles of the same subject from a single prompt. Perfect for product photography, character design, or any project where you need consistent subjects from different perspectives—all in one command.

What is the Creative Director tool?

Creative Director is your AI creative partner that generates complete creative sets or variations for products and characters in one command. Instead of creating assets one by one, get an entire cohesive set of images that work together—perfect for marketing campaigns, branding, or product launches.

Can I train custom AI models on my own content?

Yes! Our Training Lab lets you train custom models on your characters, styles, or products. Generate with Flux or Krea using your trained models, combine multiple trained models, use reference images, inpaint, and create with image prompts. We also offer pre-trained Kolbo models for specific styles like iPhone photos, storyboards, and vintage aesthetics.

Can I compare results from different AI models?

Yes! Our Multi-Model Generation feature lets you send the same prompt to multiple AI models simultaneously and compare results side-by-side. This is different from Auto Select—it's about testing different models to find the perfect output for your needs. Currently available in Kolbo.AI, Weavy, and Pollo AI.

What is the Auto Select Model feature?

Auto Select Model is our AI agent that automatically chooses the best model or tool for your specific request at every step. You don't need to worry about which AI model to use or compatibility issues—Kolbo.AI intelligently selects the optimal option based on your task, ensuring you always get the best results without manual configuration.

Can I run multiple AI generations at once?

Yes! Kolbo.AI supports unlimited concurrent generations, meaning you can run multiple AI tasks simultaneously without waiting. Start several image generations, video creations, or other tasks at the same time and they'll all process in parallel. This feature is also available in Weavy.

Can I generate 3D models?

Yes! Kolbo.AI supports 3D model generation from text descriptions or images using AI. This feature is only available in Kolbo.AI and Krea, making it a rare capability in the AI creative space. Perfect for game dev, product visualization, or creative projects.

How do I generate images?

Generate high-quality images using 20+ leading AI models including Seedance, Flux, Reve, GPT, Ideogram, Nano Banana, Qwen, and Grok Imagine. Use the Training Lab to create custom models, Canvas Tool for AI + manual editing (upscale, faceswap, reframe, remove background, add text, Lightroom-style adjustments), Angels for multi-angle shots, or Creative Director for complete creative sets. All results reflect your project's context and memories for maximum relevance.

Can I create videos in Kolbo.AI?

Yes—comprehensive video creation with multiple approaches: (1) Text-to-Video: Use Kling, Hailuo 2.3, Seedance, Sora 2, or Veo 3 with preset effects and camera movements. (2) Image-to-Video: Animate images with First Last Frame, or use Motion Transfer to apply reference video movements to your characters. (3) Advanced: Lipsync for dubbing (image-to-video or video-to-video), Character Replace to swap characters in videos, Elements to add and animate multiple objects together. (4) Editing: Built-in video editor with upscale to 4K, reframe, remove background, and audio generation. All within your project session.

Does Kolbo.AI have video editing tools?

Yes! We include a free internal video editor with professional capabilities: upscale videos to 4K, reframe compositions, remove backgrounds, generate audio, and more. Plus, with our Adobe Premiere/After Effects plugins, you can seamlessly integrate AI-generated content into your existing editing workflow.

How does Kolbo.AI handle audio and music?

Comprehensive audio tools: (1) Music: Generate custom music with Suno from text prompts. (2) Voice: Text-to-speech with Eleven Labs and DeepDub, including voice cloning to create custom voices. (3) Transcription: Upload audio or video files to chat for instant transcription and subtitle creation. (4) Sound Effects: Text-to-sound for custom audio effects. All audio tools integrate with your projects and can be analyzed in chat.

How does project management work in Kolbo.AI?

Each project is a complete workspace containing all your related work—sessions from every type of AI tool—organized in one place. Projects support: (1) Shared context across all AI models and tools. (2) Organized sessions for chat, images, videos, audio, and more. (3) Memories that persist across the entire project. (4) Team collaboration with full session and project sharing. (5) Folders and organization. (6) Quick search across all content. (7) Export to PDF and other formats. You always have a clear overview, and nothing gets lost.

How does the memories feature help me?

Kolbo.AI's memories let every AI tool in the platform access your project's history and essential information. This means your content is never scattered or forgotten—you can reuse and build on past work across any tool or session. You can even import memories from ChatGPT. Memories ensure continuity and context across your entire creative workflow.

What is the Artifacts Window?

The Artifacts Window is a powerful feature for creating and sharing visual code and documents. Generate landing pages, websites, documents, and other structured content that displays in a separate window. These artifacts are shareable as standalone sites—perfect for email templates, website mockups, presentations, or documentation. It's like having a built-in code editor and preview system.

How does team collaboration work?

Kolbo.AI offers comprehensive team collaboration: invite team members, assign roles, share full sessions and projects (not just results), work in real-time, manage feedback, and collaborate with complete workspace access. This level of deep collaboration—sharing entire sessions with context—is only available in Kolbo.AI and Higgsfield. Plus, use the Prompt Library to save and share templates across your team.

Can I analyze files like audio, video, or large documents in chat?

Yes! Simply upload audio, video, or document files directly to any chat session for instant analysis. Our AI can transcribe audio/video, summarize documents, extract insights, generate subtitles, or answer questions about the content. This "insane feature" makes Kolbo.AI perfect for content analysis and research workflows.

Are there usage limits?

Yes, each plan has clear usage tiers based on credits. All limits are transparent, and your current usage is always visible in the dashboard—no surprises. With our Credits Roll Over feature, unused credits accumulate monthly as long as your subscription is active.

Can I export my content?

Yes, easily export any text, image, video, audio, 3D model, or entire projects whenever you want. Export to PDF, download media files, or save complete project archives. Your content is always accessible and portable.

How is my data protected?

Your privacy is our priority. Data is private by default and never used for AI training unless you give explicit permission. Even our AI vendors (like OpenAI, Anthropic) are required by contract to delete your data within 30 days. If you don't grant permissions, your data remains fully private, secure, and encrypted. We follow enterprise-grade security standards to protect your creative work.

How do I get support or suggest new features?

Our support team is always available through the platform. Reach out via in-app chat, email, or social media (@kolbo.ai). Feedback and feature requests are welcomed—we're actively building Kolbo.AI together with our community and regularly implement user suggestions.

Change Log

Track all the updates and improvements we've made to Kolbo.AI.

June 4, 2026

ORGANIZATION & WORKFLOW

Cleaner mobile Create menu: tapping a category (like Videos) on the mobile bottom nav now replaces the category grid with the sub-tools and a back button, so the full list of tools is always visible — even on the narrowest viewports. The sub-tools layout also adapts smarter, using a clean 2×2 grid when a category has exactly four tools.

IMPROVEDBetter fit inside the Adobe plugin: minimized prompt panels, message bubbles, and the mobile bottom-nav menu now render correctly at the Adobe CEP plugin width. Previously, panels in that width band would lose their glass look and the bottom-nav sub-tools could be cropped above the viewport.

FIXEDAdobe plugin scrolling: fixed an issue on Mac where mouse-wheel scrolling inside the Kolbo Adobe plugin would stop working after clicking another panel or switching apps, requiring a refresh to recover. Scrolling now stays responsive without needing to refresh the panel.

Stuck-overlay recovery: occasionally a closed dialog or selector could leave an invisible layer behind that blocked wheel and scroll input across the whole app. Kolbo now detects and clears these stray layers automatically.

May 31, 2026

NEW MODELS

NEWGemini Omni Flash (image-to-video): Google's new multimodal video model accepts text, image, video, or voice as input to create coherent video with real-world logic. Available in the Elements tool under image-to-video.

May 24, 2026

DEVELOPER API (SDK)

Route any generation into a specific project: Every generation endpoint in the Developer API and MCP server (image, image edit, video, video-from-image, video-from-video, music, speech, sound, chat, creative director, elements, first-last frame, lipsync, transcription, 3D, and the global image / video edit routes) now accepts an optional project_id body field. When set, the generated asset lands inside that project instead of the auto-created 'API Generations' bucket. Useful when your scripts or MCP-driven agents work across multiple Kolbo projects and you want results to show up in the right one.

NEWNew GET /v1/projects endpoint: Lists the projects your API key can write to, each with its id, name, your role on it (owner / full / edit), and a flag marking the default 'API Generations' project. Lets LLM agents driving the MCP resolve a project name to an id before posting a generation.

ORGANIZATION & WORKFLOW

Reach the bottom of every list on mobile: Across every generation tool (image, video, audio, training lab, creative director), the last items in your generation list and the empty-state content used to slide under the prompt panel and become unreachable when the panel expanded. Each tool now sizes its bottom spacing to the live height of the prompt panel, so the last result you generated is always tappable no matter how tall the panel grows.

Mobile sheet menus no longer trigger the thing behind them: Dismissing a mobile bottom sheet (image actions, model picker, voice picker, and the rest) used to let the same tap fall through and accidentally select an image card or open a preview underneath. Sheets now absorb the dismissing tap so you only ever close the sheet.

Smoother resize behavior for the mobile prompt panel: The recalculation that runs while the collapsible prompt panel expands or collapses is now batched per frame, which removes the small visible jumps in content padding during the open/close animation.

May 22, 2026

NEW MODELS

NEWGemini Omni Video: A new text-to-video and elements model that can pin specific character voices and audio clips into the result. Supports 4–10 second outputs at 16:9 or 9:16 in 720p, 1080p, or 4K. Reuses the voice samples attached to your character Visual DNAs — say it once on the DNA and every Gemini Omni run picks up the right voice automatically.

Grok Imagine Quality (text-to-image and edit): xAI's higher-fidelity Grok Imagine tier is now available as both a text-to-image model and an image-edit model, with selectable 1K and 2K output. Supports Visual DNA, Cinema mode, and up to 3 reference images per edit.

Flux Pro Erase: A premium dedicated erase model from Black Forest Labs, selectable in the Inpaint tool's Erase mode. Use it when you want cleaner removals on tricky surfaces (skin, hair, complex textures) than the default erase model produces.

NEW30 new Gemini text-to-speech voices: A whole new family of 30 expressive Gemini voices is now available in Text-to-Speech, each with 40+ language and accent variants. Pair them with the new Voice direction picker and Accent selector to dial in exactly how a line is delivered.

IMAGE FEATURES

Attach reference video and a voice sample when creating a Visual DNA: The Visual DNA creation panel now includes an Advanced media section where you can optionally attach a short 3–10 second video clip and, for character DNAs, a voice sample alongside the usual reference images. Useful when motion or speech is part of what makes the character or world feel right — captured once at creation time instead of only after the fact.

NEWNew Erase mode in the Inpaint tool: The brush tool now has two modes — Replace (paint a region and describe what should appear there) and Erase (paint a region and have it cleanly removed, no prompt needed). The model list swaps automatically when you switch modes, and a new Inpaint / Erase shortcut is available from the image right-click menu so you can jump straight into either mode from any result.

AUDIO & MUSIC

Voice direction picker for Google / Gemini voices: When you pick a Gemini voice you'll now see a Voice direction dropdown with 12 curated styles — Warm & conversational, Dramatic newscast, Whisper, Excited & energetic, Calm narration, Cheerful, Serious & professional, Storyteller, Sad & melancholic, Soft & intimate, British accent, and TV commercial — plus a Custom option for free-form direction up to 500 characters. The chosen style is shown on the generated audio card so you can see exactly how each take was directed.

NEWAccent selector for Google / Gemini voices: A new Accent dropdown sits next to the Voice direction picker and lets you pin the voice to one of 40 locales — including English (US / UK / Australia / India), Spanish (Spain / Mexico / Latin America), French (France / Canada), Portuguese (Brazil / Portugal), Hebrew (Israel), Arabic (Egypt), Hindi, Japanese, Korean, Chinese (Mainland / Taiwan), and 20+ more. Auto remains the default for the model's own language detection.

May 21, 2026

IMAGE FEATURES

Accurate generation countdowns at higher resolutions: The countdown shown while a generation runs now reflects the actual time the chosen resolution takes. Previously, 4K image generations showed a short ~38-second timer that ran out long before the image was ready — countdowns are now resolution-aware and match real wall-clock time across image, video, and audio generations.

Clearer message when content moderation rejects a result: When a model's safety checker rejects the output of a generation, you now see an honest "Content moderation rejected the output — try a different prompt or reference images" message instead of the misleading "result expired". No more guessing why a generation silently failed.

ORGANIZATION & WORKFLOW

NEWNew API tab in Media Library: A new API section in the media sidebar collects everything you've generated through the Kolbo SDK or MCP in one place, separate from your in-app work. Useful for developers who want to browse what their scripts and agents have produced without it mixing into the rest of their library.

May 19, 2026

PLANS & PRICING

Refreshed monthly plans with more credits: Lite is now $24/mo with 2,500 credits (up from 2,250), Standard is now $40/mo with 4,250 credits, and Professional is now $79/mo with 9,500 credits. Annual plans get the same credit bumps at a 20% discount. Every plan now includes more headroom for image, video, and audio generation than before.

NEWNew Professional 2x and 3x tiers for heavy users: A new tier slider on the Professional plan lets power users scale up in one click — Pro 2x at $152/mo for 19,000 credits, and Pro 3x at $220/mo for 28,500 credits. Annual variants available at the same 20% discount. Built for studios, agencies, and creators running large generation batches who were previously stacking credit packs on top of Pro.

NEWExisting subscribers keep their current price: If you're already on Lite, Standard, or Professional, you stay on the price you signed up at — your subscription is unchanged and renewals continue at your original rate. The new pricing only applies to new checkouts and to plan changes you make yourself.

Bigger credit packs with bonus credits: The Large pack now bundles 10,500 credits and the Ultra pack now bundles 22,000 credits, both reframed with explicit +10% bonus credits over the headline amount so you can see exactly what you're getting on top of the base.

Cancel a scheduled downgrade by re-upgrading: If you've scheduled a downgrade for your next billing period and change your mind, clicking Upgrade on your current plan now cancels the scheduled change and keeps you on your current tier — no more waiting for support or letting the downgrade go through.

May 16, 2026

DEVELOPER API (SDK)

Media Library endpoints: The Developer API now exposes the full media library — list and fetch your uploads, create / rename / delete folders, add or remove items from folders, favorite and unfavorite, move items between folders, soft-delete and restore, bulk operations, and folder sharing. Everything you can do with media in the app is now available from your own scripts and MCP tools.

NEWQuick-share for HTML artifacts: A new endpoint lets you publish an HTML artifact and get back a shareable kolbo-hosted link in one call — useful for agents that produce dashboards, presentations, or landing pages and need to hand the user a URL.

Richer model metadata in list_models: The /v1/models response now includes the fields the API needed to surface but was hiding — visual_dna support, first-and-last-frame support, supported aspect ratios per generation type, default aspect ratio, min/max resolution, min/max video duration (for lipsync-video and video-to-video source clips), min/max recording duration for audio, prompt length limits, images per request, and max file size. Tooling can now build accurate UIs from a single model query.

DESKTOP APP

One-click desktop sign-in: When the Kolbo desktop app opens the approval page in your browser, it now deep-links straight to the prefilled approval screen — if you're already signed in to app.kolbo.ai, you just click Approve. No more copying device codes by hand.

IMPROVEDFaster sign-in for desktop & plugins: Already-signed-in desktop clients and Adobe / ACP plugins can now mint their Kolbo Code API key directly, without the browser device-code hand-off. The browser flow is still there for terminals and first-time setups.

ORGANIZATION & WORKFLOW

Drag a whole folder to upload: The media library now accepts folder drops — drag any folder from your desktop onto the library or a folder view and every file inside uploads in one go. Up to 200 files / 100MB per drop. Uploads land directly into the folder you dropped onto, and the view refreshes immediately so you can see what landed.

Board navigation carousel: Magic Canvas's board switcher was rebuilt as an animated, viewport-aware carousel with hover tooltips and a glass ring treatment in both light and dark mode — easier to scan dozens of boards at a glance.

NEWSequential board names: New canvas boards are now numbered sequentially and locked to that name, keeping the carousel and exports consistent across collaborators.

Hide SDK / API usage in History: Subscription → Usage History now has a toggle to hide deductions made by the Developer API or MCP tools, so you can see just the credits you spent inside the app. The preference is remembered per device.

IMAGE FEATURES

Storyboard captions: Storyboard Grid Creator now supports a caption per cell with a single toggle to turn the caption strip on or off. Captions are baked into the exported JPEG and the Manual Editor payload, with full RTL support for Hebrew and Arabic.

NEWOpen Storyboard in AI Board: A new "Open in AI Board" action sends your storyboard straight into Magic Canvas as an editable board — every image, number strip, and caption strip lands as a separate object you can rearrange, annotate, or continue editing.

CHAT & AI

Numbered media chips in chat replies: When the assistant references media you've sent, it now renders @Image1 / @Video1 / @Audio1 as color-coded chips inline — numbered per content type, so you can tell at a glance which image or clip the reply is talking about.

May 12, 2026

NEW MODELS

NEWGrok 4.3 — new xAI flagship in chat: xAI's most advanced model is now available in the chat model picker and is the new default Smart Select target for Grok. It accepts text and images, runs on a 1M-token context window, supports configurable reasoning effort, function calling, and structured outputs — strong on instruction following and low hallucination for long, agentic conversations.

IMAGE FEATURES

Aspect ratio chip on every generation: Image and video results now show the aspect ratio (16:9, 9:16, 1:1, and so on) as a small pill right next to the resolution. Visible across every image and video tool, including Creative Director batches, so you can tell at a glance which framing each result was generated at.

NEWStoryboard Grid Creator: A new tool under Creative Director for composing multiple images into a single numbered storyboard. Pick from nine preset layouts (2×2 up to 4×4) or build a fully custom grid up to 10×10, choose the output canvas ratio (16:9, 1:1, or 9:16) and a separate per-cell ratio, set a black or white background, and optionally bake in shot numbers. Fill cells from your media library, drag images straight from the side panel, or upload new ones — and drag any filled cell onto another to swap them. One-click toggle to set every image to Fill or Fit at once. Export resolution selectable from 1080p all the way up to 16K, or save the finished grid straight into your generated-images library.

ORGANIZATION & WORKFLOW

Magic Pad paper mode: Magic Pad and Magic Canvas in light mode now render as a single clean sheet of paper — flat white surfaces, no glass blur, a tighter reading column, and a subtle accent line above the AI prompt so it stays distinct. Dark mode is unchanged.

CHAT & AI

Presentations, landing pages, and interactive HTML in chat: Ask the chat for a slide deck, a landing page, or any visual interactive output (dashboards, data viz, mini-games, animated components, UI mockups) and it now switches into a dedicated artifact mode for that request. The chat detects what you're asking for in any language — English, Hebrew, Spanish, French, Chinese, Korean — and produces a single self-contained HTML artifact built specifically for that format: viewport-locked slides with arrow-key navigation for decks, hero + features + CTA structure for landing pages, and previewable interactive components for everything else. A small badge appears on the message so you can see which mode the chat picked.

May 11, 2026

NEW MODELS

Topaz Labs video upscaling: Upscale your videos to 4K with the official Topaz Labs models — the same engine professional studios use for high-end finishing.

NEWMidjourney V8 and Niji V7: Two new Midjourney models are now available — V8 for the latest aesthetic update and Niji V7 for stylized anime and illustration. Each run delivers four images.

IMAGE FEATURES

NEWBackground remove and replace: New one-click background removal and AI-driven background replacement, powered by Ideogram.

NEWFlux2 and Flex Max image editing: Two new high-fidelity image editing models added for precise instruction-based edits.

ORGANIZATION & WORKFLOW

Magic Note version history: Magic Pad and Magic Canvas now automatically keep a rolling history of recent versions. Open the version list to preview any earlier state and restore it — the current state is snapshotted first, so a restore is always reversible.

Live generation ETA: Image and video generations now show a live countdown so you can see roughly how long each run will take. The estimate appears once the run has actually started and adapts to the model you picked — no fabricated time before the server confirms.

Visual DNA recently used: The Visual DNA picker now surfaces a Recently used row at the top, so the casts and characters you reach for most often are one click away.

Move sessions between projects: Any session — chat, image, video, audio, magic note and more — can now be moved into a different project from its menu. You'll only see destinations where you have edit access, and shared collaborators with view-only access cannot have sessions pushed into their projects.

May 4, 2026

PAYMENTS & BILLING

Quick Buy: Purchase credit packs with a single tap using your saved card — no Stripe redirect, no checkout page. A 2-second safety countdown prevents accidental charges.

Promo codes in Quick Buy: Enter a promo code directly in the Quick Buy confirmation modal — see the discount applied live before you confirm.

Direct plan upgrades: Switch subscription plans instantly without going through the billing portal — upgrade or downgrade in one click.

Retention offer: If you're considering cancelling, a 30% discount is automatically offered on your active subscription.

GENERATION RELIABILITY

Credits protected on cancel: Cancelling a generation now fully protects your credits — even if the AI provider finishes processing after your cancel request, credits are never deducted.

Smarter retries: Transient network errors are automatically retried with backoff so temporary connectivity issues don't silently fail your generations.

IMPROVEDImproved generation recovery: Better detection and automatic restart of generations that stall mid-process.

MOBILE UX — VIDEO TOOLS

More compact mobile controls: Preset, Visual DNA, and Motion Preset selectors are now inline with the model chip row — saving a full row of vertical space on mobile.

iOS scroll fix: The last items in bottom sheets (duration picker, model lists) are no longer hidden under the gesture bar.

PROMPTS

Click to copy: Click any prompt in your generation history to copy it instantly — works across image, video, and editing tools with RTL support.

DEVELOPER API (SDK)

NEWNew SDK endpoints: Image editing and video editing are now available through the Kolbo Developer API.

SUPPORT

IMPROVEDSmarter support tickets: AI-assisted ticket creation now includes your full conversation transcript, tool usage trail, and error context — so support can resolve issues faster.

GENERAL

FIXEDOverall bug fixes and stability improvements.

April 26, 2026

NEW CHAT MODELS

GPT-5.5 (OpenAI): OpenAI's latest flagship model — 1M context window, 128K output, reasoning effort control, and vision support.

Claude Opus 4.7 (Anthropic): Anthropic's most capable model — advanced reasoning and long-horizon agentic tasks.

Grok 4.20 (xAI): 2M context window, lowest hallucination rate, strict prompt adherence, and built-in reasoning.

Gemini 2.5 Pro & Flash (Google): Google's latest Gemini models — 1M context, multimodal (video + audio), advanced reasoning. Flash is ultra cost-efficient.

Gemini 3 Pro (Google): Google's newest and most capable Pro model — top-tier reasoning and multimodal understanding.

GLM-5.1 (Z-AI): Advanced coding and long-horizon autonomous tasks model with 200K context.

MiniMax M2.7: Multi-agent collaboration model for financial modeling, document generation, and complex workflows.

NEW IMAGE MODELS

GPT Image 2 (OpenAI): OpenAI's best image model now available in Text-to-Image and Image Editing. Supports up to 16 reference images, 7 aspect ratios, and Visual DNA.

NEWFlux 2 Flex & Max (Black Forest Labs): Two new Flux 2 image generation models — Flex for flexible resolutions, Max for premium quality. Both support Visual DNA and up to 2K resolution.

ImagineArt 2.0: Next-gen photorealistic image model with 14 aspect ratios including 21:9 cinematic and banner formats.

NEW VIDEO MODELS

Kling V3 4K & O3 4K (Kuaishou): Native 4K video generation — text-to-video, image-to-video, and reference-to-video with multi-shot support, native audio, and Visual DNA character consistency.

Wan 2.7: Full suite across all video tools — text-to-video, image-to-video, Elements, and video-to-video. Supports audio input and Visual DNA.

Veo 3.1 Lite (Google): Cost-efficient Veo 3.1 with native audio generation — available in text-to-video, image-to-video, and first-last-frame.

Grok Extend Video: Extend any video using Grok's generation engine — with prompt guidance for the continuation.

LTX 2.3 Extend Video: Extend videos from the start or end with flexible duration control.

NEW AUDIO MODELS

Stem Splitter: Split any track into individual stems — vocals, drums, bass, guitar, keyboard, strings, brass, and more. Download each stem separately.

Vocal Separator: Cleanly isolate vocals from any audio track.

IMPROVEDMiniMax Music upgraded to v2.6: Higher quality music generation with instrumental mode support.

VOICE & TTS REDESIGN

MAJORVoice Cloning redesign: Completely rebuilt voice upload flow with support for 41+ languages (with country flags), duration requirements per provider, image reference upload for character-based voices, and prompt audio input for MiniMax.

Voice Design: Upload a character image to influence AI voice creation — the model uses the visual reference to shape tone and style.

NEWVoice Selector: New comprehensive filtering system — filter by language, gender, provider, character type (narrator, wizard, villain, robot...), and use-case. Animated chip interface.

AUDIO TOOLS

Separated Tracks Viewer redesign: Play, control volume, and download each stem individually from the music generation results.

Music Generator: Live prompt length counter with color-coded warnings. Auto-expanding input, collapsible prompt area, and session persistence.

NEWMusic Presets: New chip UI shows the selected preset thumbnail inline in the prompt. Only one preset plays at a time.

NEWSeedance 2: 200+ new video presets added.

Audio trimmer in video editor: Trim audio clips from your Media Library directly inside the video editor.

TRAINING LAB

Model card redesign: Drag horizontally on any model card to adjust its blend strength (0.1–1.5) — live visual indicator and tooltip.

MAJORTraining progress redesign: Multi-stage progress tracker — Analyzing → Preparing → Training → Refining → Finalizing → Complete — each stage with animated icon and rotating status messages.

FLOW BUILDER

IMPROVEDMajor node overhaul: Crop, Trim, Extract, Download, and Speed nodes completely rebuilt with better layout and reliability.

Zoom-adaptive rendering: Nodes automatically simplify at lower zoom levels for a cleaner overview.

Visual DNA in Flow: Use Visual DNA characters directly inside Flow generation nodes.

Stable thumbnails: Connected media thumbnails stay in sync after re-generation.

Smart-select model support: Flow nodes can now route through gateway and smart-select models.

IMAGE TOOLS

Image Preset Modal redesign: Dual-mode display — switch between generation presets and image editing effect presets. Category filtering and masonry/grid layout toggle.

Model selection for Face Swap, Remove Background, Replace Background, and Convert SVG — choose the best model for each task.

VIDEO TOOLS

NEWExtend Video: New tool to extend any video at the start or end with an optional prompt guiding the continuation.

VISUAL DNA

Use Visual DNA without reference images: Creative Director now accepts Visual DNA alone — no extra uploads required.

FIXED@mention in scene prompts: Tag characters by @name in Creative Director scenes — automatically resolved to the correct reference images.

MAJORVisual DNA chip display: Video generations now show which Visual DNA presets were used, with one-click reuse. Redesigned chip styling.

CREATIVE DIRECTOR

Multi-shot / multi-clip toggle: Supported models can generate multi-shot sequences from a single brief.

Motion Graphic reuse: Select any previous motion graphic and reuse its prompt directly to iterate on it.

HELP WIDGET

AI Support Agent: The help widget is now powered by a full Claude AI agent — context-aware answers based on which tool you're on.

Image attachment: Drag and drop screenshots directly into the help chat.

Guided ticket creation: AI walks you through filing a support ticket, auto-filled with your conversation context.

PROJECT CREATION

NEWRedesigned project dialog: New tabbed sidebar — Details, Members, Files, and Memory. AI-powered project description generation from your project name and uploaded files.

NEWNew project selector: Searchable project switcher available anywhere in the app without leaving your current tool.

ACCOUNT

30-day account deletion grace period: When you delete your account a restore link is sent by email — permanent deletion only happens after 30 days.

Account restore page: Dedicated page to recover your account from the email link before the grace period expires.

NEWSupport Tickets: New dedicated page to submit and track your support tickets — see status updates, admin responses, and ticket history in one place.

ORGANIZATIONS

Org portal sidebar: Full navigation sidebar for all org roles.

12 languages across the entire org portal with a language selector and light/dark theme toggle in the header.

Collaboration email notifications: Automatic emails for org invitations, project shares, and member changes.

DEVELOPER API (SDK)

NEW9 new generation types added to the SDK: Elements, Lipsync, First-Last Frame, Video-from-Video, 3D generation, Transcription, Media upload, Media library, and Presets.

KOLBO CODE

Kolbo Code is now live — a CLI tool that brings the full Kolbo.AI platform into your terminal and IDE. Generate images, videos, audio, and more without opening a browser.

Kolbo Code MCP: Connect Kolbo to Claude Desktop and other AI tools via MCP — use all Kolbo generation capabilities directly from your AI assistant.

No Node.js required: Install in one command on Mac (curl) or Windows (irm).

GENERAL

FIXEDOverall bug fixes and stability improvements.

April 7, 2026

NEW MODELS: MIRAGE FAMILY

NEWIntroducing MIRAGE — 4 custom AI models trained in-house by Kolbo, exclusive to the platform.

MIRAGE FILM: Cinematic film aesthetic with grain, depth, and mood. Every frame looks like a high-budget production still. 2 credits.

MIRAGE FILM Ultra: The premium version built on Flux 2. Richer detail, more accurate lighting, director-level output. 6 credits.

MIRAGE LIFE: Hyper-realistic lifestyle photography in the style of iPhone influencer content. Natural light, authentic moments, social-ready. 3 credits.

MIRAGE CAST: Hybrid character pipeline — generates full-body shots and face close-ups from one prompt for consistent characters across every frame. 4 credits.

P Image Upscale: Lightning-fast image upscaling to 8MP at 4x in under 1 second. 4 credits.

NEW VIDEO MODEL: SEEDANCE 2

Seedance 2 by ByteDance is now fully integrated across all video tools. Sharper motion, stronger prompt adherence, and more cinematic output than its predecessor.

CAST SYSTEM

NEW5 new character collections added: Attractive, Blonde, Curvy, Hot Girls, Hot Guys — 600+ new pre-generated character references.

Cast V2: Glamour and Street collections with AI-analyzed attributes (gender, ethnicity, body type, age range).

Outfit descriptions now supported on cast members for more detailed character control.

NEWNew Fantasy body type option for non-human characters and creatures.

VISUAL DNA

Folders: Organize Visual DNAs into color-coded custom folders with item counts.

Search & filter: Full-text search across names, tags, and descriptions. Tag-based filtering with OR logic.

Soft delete & restore: Deleted Visual DNAs can now be recovered — old generation references stay intact.

Multi-select mode: Select multiple DNAs at once across Platform, Organization, Collaborator, and Personal scopes.

UX IMPROVEMENTS

Drag-and-drop reference images across all tools — reorder, copy, and manage references with position badges and context menu.

NEWNew cancel dialog with clear credit breakdown before cancelling a subscription.

Media hover previews in billing usage rows.

Invoice search, date range filtering, and batch download.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 30, 2026

NEW FEATURE: 3D WORLDS

NEWIntroducing 3D Worlds — a completely new tool in Kolbo.AI. Build and explore fully interactive 3D environments directly in your browser. Powered by Gaussian Splatting and WebGL 2.0 — no software installs required.

IMPROVEDNavigate freely with a virtual camera, fly through scenes, and capture AI-optimized snapshots from any angle.

Google Maps to 3D: Import any street view from Google Maps and Kolbo.AI instantly converts it into a navigable 3D environment.

Sketchfab integration: Pull any 3D model from Sketchfab directly into your scene.

3D objects & mannequin poses: Add and position 3D objects in your scene, including poseable mannequins for character references.

Camera controls: Multiple focal lengths, F-stops, and aspect ratios for cinematic composition.

Environment presets: Choose from apartment, city, forest, studio, sunset, and more to set your scene lighting.

Custom 3D model creation: Generate custom 3D models for your products or characters from scratch.

Full undo/redo, session history, and scene management.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 23, 2026

NEW FEATURE: FLOW TOOL

NEWIntroducing the completely rebuilt Flow Tool — a visual node-based workflow builder for chaining AI tools together. Connect image generators, video tools, audio tools, text generators, and more into fully automated pipelines.

NEW13 brand new node types: Elements Video, Director, Crop, Trim, Speed, Download, Extract, Merge, Group, Video Generation, List, Media Library, and AI Assistant nodes.

NEWNew canvas experience: Right-click context menus, multi-node selection toolbar, per-node run and regen menus, node formatting toolbar, and node preview dialog.

Per-node model selection: Choose which AI model runs at each step of your workflow.

Full undo/redo, copy/paste node groups, and node duplication.

Flow Preview page: Share a read-only preview of any Flow session.

NEW TOOLS

NEWNew: Thumbnail Generator — dedicated tool for generating YouTube and social media thumbnails with reference image support and live preview.

NEWNew: Ad Creatives — focused tool for generating advertising creatives and banner images.

NEWNew: Logo Maker — dedicated AI logo generation tool with style controls and brand customization.

VIDEO MODELS

NEWNew model: Grok Video — Grok's video generation model is now available across video tools.

NEWNew model: Sora 2 — OpenAI's latest cinematic video model.

NEWNew model: Veo 3.1 — Google's enhanced video generation model.

NEWNew model: LTX 2 — updated high-quality video generation.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 20, 2026

APP-WIDE UX/UI REDESIGN

NEWThe entire Kolbo.AI interface has been redesigned with a new Liquid Glass design system — consistent visual language, updated components, dark/light mode improvements, and a cleaner layout across every tool and page.

GENERATION UX

Prompts are now retained after generation — your text stays in the input field after you hit generate, across all tools. No more retyping.

AUTHENTICATION

NEWNew: OTP passwordless login — sign in with just your email using a one-time code, no password required.

Google sign-in now opens in a popup — no more full-page redirects. You land directly in the app after signing in.

MOBILE APP

IMPROVEDImproved iOS and Android builds — more reliable TestFlight and Play Store delivery.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 12, 2026

ORGANIZATION: DIVISIONS & SPACES

NEWNew: Divisions & Spaces — a new hierarchical structure for organization management. Create divisions, allocate credits per division, manage members, and set tool visibility per division.

Spaces: Sub-units within divisions for even more granular credit allocation and access control.

Division model preferences: Set which AI models are available to each division.

Division navigation control: Control which tools and sub-tools are visible to division members.

ORGANIZATION ANALYTICS

NEWNew: Organization Live Monitor — real-time dashboard showing all active generations across your org, including model usage and user activity.

NEWNew: Organization Credit Usage — detailed analytics showing credit consumption by user, tool, and time period.

MEDIA LIBRARY

IMPROVEDMajor performance overhaul: media library now loads up to 100 items per page (was 20) with virtualized rendering throughout — dramatically faster with large collections.

Model store memory optimization: reduced from 26MB to 1MB through shared identifier maps — the app is significantly lighter across all tools.

Global Media Picker: access your full media library from anywhere in the app without leaving your current tool.

Variation support: generate variations of any image directly from the media grid.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 11, 2026

AUDIO & MUSIC

NEWNew music provider — Sonauto: Generate AI music with Sonauto v3 (variable-length, 2–4 min, high quality) and Sonauto v2 (~1:35 min with extend support — upload audio to keep going).

Extend any song: Upload existing audio and extend it with Sonauto v2.

CHAT

IMPROVEDImproved folder organization with faster sidebar loading.

IMPROVEDImproved import interface for bringing in conversations from Claude and ChatGPT.

USAGE & BILLING

Clearer credit usage breakdown across generation types.

GENERAL

FIXEDOverall bug fixes and stability improvements.

March 10, 2026

CHAT IMPORT FROM AI PLATFORMS

Import from Claude & ChatGPT: Upload your data export from Anthropic Claude or OpenAI ChatGPT to bring all your conversations into Kolbo. Memories are extracted automatically.

Auto platform detection: The system automatically detects whether your export is from Claude or ChatGPT and parses it accordingly.

Import folder: All imported conversations are organized into a dedicated folder named with the platform and import date.

Memory import: Memories from your Claude export are extracted using AI and saved to your Kolbo memory store with duplicate detection.

NEWQuick access: Import button added directly to the Chat sidebar — hover the upload icon next to New Chat to import chats or memories.

SDK & DEVELOPER API

MAJORCreative Director API: Generate complete creative sets programmatically via the Developer API with multi-scene status tracking.

NEWImage-to-Video API: New endpoint for generating videos from images through the SDK.

VIDEO EDITOR

NEWRemotion enhancements: New transition effects and improved export presets for motion graphic generation.

INTERNATIONALIZATION

Chat import translations: Full i18n support for the chat import feature across all 12 languages.

March 9, 2026

NEW MODELS

GPT-5.4: OpenAI's most capable frontier model with 1.05M context window and 128K output tokens. Supports reasoning effort control, verbosity settings, vision, streaming, and tool use.

IMPROVEDLTX-2.3: Updated video model now available across video generation tools with improved quality.

IMPROVEDKling 3.0: Kling 3.0 Standard and Pro now available with faster and more reliable video generation (text-to-video and image-to-video). Includes multi-shot scene support, Visual DNA character elements, and built-in audio.

SHARED CREDITS IN PROJECTS

Shared billing: When working in a shared project, credits are now charged to the project owner's account. All generation types support shared billing — image, video, audio, music, chat, speech, elements, training, 3D, motion graphics, and more.

Clear error messages: If the project owner runs out of credits, collaborators see specific messages explaining the situation instead of generic errors.

TELEGRAM & WHATSAPP BOTS

Telegram bot (@kolboai_bot): Generate images, chat with AI, edit photos, and check credits directly from Telegram. Supports voice/video/audio analysis, Google Search grounding, and smart intent detection.

WhatsApp bot: Same AI capabilities available on WhatsApp via Meta Cloud API — image generation, chat, and media analysis.

CONTENT SAFETY

NSFW output detection: AI-generated images and videos are now scanned for NSFW content. Flagged outputs are replaced with a placeholder image. Configurable per model.

GENERATION RECOVERY

Unified recovery system: If a generation gets interrupted (server restart, timeout, network issue), the system now automatically detects and resumes it. Works across all providers and all generation types.

Retry storm fix: Terminal failures (e.g., content policy violations) are no longer retried endlessly.

CREDIT SYSTEM & ADMIN

Credit bucket breakdown: Admin panel now shows plan credits, purchased credits, and redemption credits separately instead of a single total.

Redemption-only editing: Admin credit adjustments target redemption credits only — plan and purchased credits are protected from accidental modification.

Inline credit editing: Quick credit adjustments directly from the admin user table with full bucket breakdown display.

PROMOTIONS

Countdown timers: Promotional badges and banners now show live countdowns (dd:hh:mm:ss) when the promotion has an end date. Visible on the navigation bar crown badge, pricing page, and promotion banners.

MEDIA LIBRARY

MediaImportDialog refactored: Import dialog rebuilt as a thin wrapper around the core media library — consistent behavior everywhere, fewer bugs, simpler codebase.

Global media panel simplified: Removed duplicate controls; now uses the same header and project selector as the main media library.

CHAT & IMAGE EDITING

FIXEDImage editing in chat: Fixed in-chat image editing workflow — reference thumbnails now show image previews instead of file icons.

OpenAI streaming fix: Regular chat now uses the Chat Completions API for full streaming support. The Responses API is only activated when web search is explicitly requested.

VIDEO & EMBEDDED VIEWS

Video container fixes: Videos now properly fill their container in Electron and Adobe CEP embedded views.

FIXEDResizeObserver fix: Fixed animation glitch caused by ResizeObserver loop in embedded view.

SDK & API

SDK expansion: Music generation, text-to-sound, and text-to-speech now available through the Developer API.

NEWLTX-2.3 support: Proper duration handling and model options parsing for the new video model.

PLATFORM IMPROVEMENTS

Visual DNA & Moodboards: Now accessible to all members in shared projects.

Kling O1/O3 image editing: Automatic image resize when reference images exceed 10MB file size limit.

Storyboard prompt: Minimum duration lowered from 5 seconds to 1 second for more granular scene control.

Fallback routing: Sound and audio settings now carry through correctly when a generation falls back to an alternative model.

Subscription messaging: Free plan users see clearer credit refresh dates; purchased and gift credits show 'never expire' wording.

Bot detection fix: Credit pack purchasers are no longer incorrectly flagged as bots.

IMPROVEDGeneration cancellation: Improved cancellation handling across elements, image, text-to-video, and video controllers.

March 5, 2026

NEW MODELS

Kling V3 Motion Control Pro: High-fidelity motion transfer from reference video to character image. 26 credits/sec. Supports image/video orientation, optional prompt, audio preservation, facial consistency elements.

Kling V3 Motion Control Standard: Cost-effective motion transfer from reference video to character image. 19 credits/sec. Same features as Pro at a lower price point.

March 4, 2026

NEW PLUGIN

Kolbo Studio for Adobe: Full Kolbo.AI platform now runs natively inside Premiere Pro and After Effects. Generate images, video, voiceovers, music, and captions directly in your timeline. AI captions, silence removal, and AI music placement built in. Available for Windows and Mac. 150 free credits on install.

NEW MODELS

NEWQwen Image 2 Standard: New text-to-image and image editing model. 5 credits per generation. Supports up to 3 reference images and 5 aspect ratios (1:1, 4:3, 3:4, 16:9, 9:16).

Qwen Image 2 Pro: Higher-quality Qwen image model for text-to-image and image editing. 11 credits per generation. Same reference image and aspect ratio support as Standard.

February 27, 2026

NEW MODELS

IMPROVEDImage: Nano Banana 2 — upgraded model for image editing and text-to-image with faster generation, improved photorealism, wider stylistic range, and better prompt adherence. Available in Image Tools → Text to Image and Image Editing. Web search toggle for realtime results based on the internet.

STABILITY & PERFORMANCE

IMPROVEDHelp widget and chat tools: Better stability across help widget and chat tools

FIXEDBug fixes across the app

IMPROVEDServer upgraded for better speed

TRANSCRIPTION & SUBTITLES

Transcription and auto subtitles: Support for no pronunciations and stretch captions (toggle on/off)

PLUGINS

NEWAdobe plugin: New version with better stability and new captions options

February 24, 2026

AI MODELS & INTEGRATIONS

ByteDance Dola Seed 2.0 Mini: Advanced multimodal AI for video, image, and audio content analysis

Training Lab Model Expansion: Support for multiple base models (flux.2, qwen, krea, wan, z-image) loaded dynamically from backend for maximum creative flexibility

IMPROVEDSeeDream 5.0 Lite: ByteDance's latest image generation model with enhanced prompt understanding and intelligent reasoning

VIDEO FEATURES

IMPROVEDVideo Editor - Batch Operations: Select and manipulate multiple clips simultaneously for professional editing workflows (25% faster than sequential edits)

NEWVideo Editor - New Aspect Ratios: Added 21:9 ultra-wide cinematic and 4:5 vertical social media formats

Video Editor - Cinema & Full Screen Modes: Immersive editing experience with distraction-free layouts

Motion Graphics with Sound: AI-generated synchronized background music and sound effects for animated videos

Real-Time Motion Graphics Editing: Instant preview of changes without waiting for regeneration (<50ms preview latency)

AUDIO FEATURES

Speaker Detection: AI identifies and labels different speakers in transcriptions (85-95% accuracy)

ORGANIZATION & WORKFLOW

Smart Project Memory: AI automatically captures and remembers project details from conversations

PDF/Text Context Upload: Extract project context from documents, brand guidelines, and specifications

Prompt Copilot - Project Filter: Filter saved prompts by current active project for quick access

Project Files Across Chats: Upload files once, access from any conversation within the project

INTERNATIONALIZATION

Instant Language Switching: Change interface language without page refresh across all 11 supported languages

PERFORMANCE & UX

Real-Time Credit Counter: Live credit balance updates via WebSocket (80% fewer API calls)

IMPROVEDAuto Model Performance: 40% faster model switching with 60% fewer component re-renders (150ms → 90ms)

Cleaner Auto Model UI: Compact badge design (70% smaller) showing icon + name only

Chat re-renders: 60% reduction for smoother performance

Stable 60fps animations across the platform

February 19, 2026

NEW MODELS

Image: Recraft 4 in Text-to-Image and Image Editing

Image: Kling O3 image generation (Standard & Pro) and Nano Banana Base

Video: Kling 3.0 Standard & Pro (Text-to-Video & Image-to-Video) with built-in audio, multi-shot, 3-15 second clips

Video: Kling O3 Standard & Pro Video Edit - edit videos with natural language using a reference image or character

Music: Google Lyria 2 (up to 30s, non-lyrical) and Minimax; Music Vision - upload an image or video to generate matching music

MAJOR FEATURES

Instant feedback: Optimistic UI—progress cards appear right away when you generate; smoother experience and less dependent on internet speed (images, video, lipsync, speech, music, sound)

IMPROVEDMagic Pad: Full redesign with improved prompt box and uploads for Magic Pad and Magic Canvas

IMPROVEDVisual DNA: Better character consistency; Main Image (best reference) required, up to 3 supplemental images; video and audio upload for voice and motion consistency; Kling voice; shareable links for moodboards and Visual DNA; organization-wide presets; works in Creative Director Cinema modes

FIXEDMusic Generator: Generate music from an image or video (Vision); choose from your Media Library; lyrics and instrumental options per model; extend tracks, presets, trim and convert audio; fixed duplicate uploads and credit handling

VIDEO EDITOR

MAJORComplete overhaul: Refined timeline with precise drag-and-drop for clips

NEWNew rendering for smoother previews and all visual effects (blur, brightness, contrast, saturation, hue, temperature, fade-in/out, zoom, and more) per layer

IMPROVEDSmoother playback with better sync between picture and sound

NEWNew timeline with thumbnails, multi-select, and snap when dragging clips

Auto-captions: Generate subtitles automatically; they appear as editable, synced clips on the timeline

MAJORRedesigned property panel: 4 tabs (Clip, Design, Effects, Layout); audio effects (fade, EQ, presets); motion presets; flip horizontal/vertical

IMPROVEDFaster in-browser export

NEWNew shortcuts: J/K/L for playback speed; Ctrl+0 to fit timeline; +/- zoom; step frame by frame

Clips no longer overlap when adding, pasting, duplicating, or moving

VIDEO TOOLS

Use @ in prompts to call Visual DNA characters (up to 3 per video); best for multi-shots and grid decoding

Kling 3.0: Best for standard realistic videos

Sound toggle for video models that support built-in audio (e.g. Kling); disabling audio can save credits (e.g. halve cost on some models)

Grok video model: Price reduced to 42 credits (down from 70-80)

Multi-shot prompts shown as visual scene cards

IMPROVEDPresets for text-to-video, camera moves, and animation style; First+Last Frame improvements; Sora 2 up to 15s and Storyboard 25s; VEO 3 extend and lower costs; Elements and Lipsync fixes; auto-captions; faster media loading

AUDIO TOOLS

NEWAudio Studio: New layout; Voice Library with emotion support; better categories with filtering and sorting; waveform view; Text-to-Speech and Speech-to-Text fixes and new models; Text-to-Sound session handling

STABILITY & PERFORMANCE

Credits: Automatic correction of credit issues; credit packs in checkout; send credits to another user

IMPROVEDSmart Select: Fixed credit reservation; better model choice for your request; supports more languages; optional auto model selection

IMPROVEDFaster and more reliable app: Better connections during generation, fewer interruptions, improved deployment reliability

IMPROVEDFaster page loads and smoother experience across the app

SECURITY & PROTECTION

IMPROVEDBot detection and protection; content filtering with user consent; improved signup checks (e.g. Gmail, academic emails); rate limiting; Google sign-in improvements; safeguards on credits and regional access

UX/UI IMPROVEMENTS

FIXEDLight Mode: All visual bugs resolved

IMPROVEDRedesigned floating nav; organization landing and billing; onboarding improvements; choose Simple or Full experience on first run; desktop fixes (black screen, dropdowns, Adobe plugin); mobile nav and credits (tap credits for usage); consistent glass-style panels; auth pages in dark mode; dashboard with Canvas shortcut and improved video loading; organization divisions with hover and Manage; subscription page default to credit packs; learning module language fallback; prompt width adapts to selected models; better wrapping for model chips

AUTH & ONBOARDING

IMPROVEDFirst-run choice: Simple or Full experience; better signup risk checks; clearer error messages for blocked regions, disposable emails, and account issues

CREATIVE DIRECTOR

IMPROVEDCinema Manual: first-last frame with cinema options; per-scene reference images for photo mode; campaign template updates; faster loading for scene assets

TRAINING LAB

FIXEDStability fixes; preset trained models library with multi-style support

COLLABORATION & SHARING

IMPROVEDShareable links for artifacts and chats; improved shared chat view; organization divisions with batch options

AI DOCS & MEDIA LIBRARY

IMPROVEDAI Docs: Now powered by Gemini and Grok; Hebrew table generation much faster and more structured

MAJORMedia Library: Redesigned with thumbnail scale slider and scroll memory—remembers where you left off when you reopen

GENERAL

IMPROVEDFaster video rendering; Adobe plugin compatibility; smarter use of AI services; improved image and video editing reliability

February 7, 2026

IMAGE TOOLS

FIXEDText to Image Combination Fix: Fixed bug where Visual DNA, moodboards, and presets didn't combine properly in text to image tool

NEWNew Grid Presets: 2x2, 3x3, and 4x4 grid presets added to image editing and text to image

NEWCharacter Sheet Preset: New character sheet preset available in text to image and image editing

FIXEDPresets Categories Fix: Fixed UX/UI issues in presets categories

NEWVisual DNA Categories: 2 new categories added - Style and Environment

FIXEDMidjourney Stability: Stability fixes for Midjourney model

FIXEDTraining Lab Fixes: Bug fixes in Training Lab

VIDEO TOOLS

Creative Director Cinema Mode: Text to video now enabled in Cinema Mode

NEWWan 2.6 Reference Model: New Wan 2.6 reference model added

NEWByteDance Motion Control: New motion control model by ByteDance

Storyboard Multi-Shot: UX/UI improvements and dialog for multi-shot creator supporting all multi-shot models

Motion Graphics Optimization: Performance optimizations for motion graphics tool

AUDIO TOOLS

FIXEDMusic Generation Fix: Fixed extract from video functionality in music generation

CHAT & AI ASSISTANTS

Context Storing: Help widget and chat tools now store context for videos and audio

IMPROVEDImproved Responses: Faster and better help widget and chat conversations with improved context handling

DESKTOP APP

FIXEDDropdowns Fix: Bug fixes for desktop app dropdowns

FIXEDAudio Trimming Fix: Bug fix in audio trimming

IMPROVEDPerformance Optimizations: Optimizations for faster and more stable web app experience

NEWFile Bridge: New feature allowing users to drag and drop and trim any audio, video, or image files from local machine - free alternative to audio library organization software

PLUGINS

Premiere Plugin Updates: Updates to Premiere plugins

February 5, 2026

VIDEO TOOLS

NEWKling 3.0: New Kling 3.0 model now available for video generation

NEWMotion Graphics Tool Launch: New motion graphics tool with enhanced presets system and improved algorithm

AUDIO TOOLS

IMPROVEDText to Speech Redesign: Voices menu completely redesigned with significantly faster loading times

IMAGE TOOLS

NEWCanvas Tool Snapping: Smart snapping feature added (like Canva) for precise element alignment

Canvas Tool Light Mode: UX/UI fixes and improvements in light mode

DESKTOP APP

NEWNew Audio Design: Refreshed audio interface design

NEWProject Selector: New project selector for better organization

NEWVideo Conversion Bitrate: Added bitrate selection for video conversion

Multi-Select with Shift: Deselect multiple items using Shift + Left Mouse Button

IMPROVEDDesktop Security Updates: Enhanced security measures for the desktop application

PLUGINS

FIXEDAdobe Plugin Updates: Bug fixes, optimizations, and UX/UI improvements for the Adobe plugin

GENERAL

MAJORMajor Security Updates: Comprehensive security improvements across the platform

Performance Optimizations: Various optimization and performance fixes throughout the app

February 1, 2026

IMAGE TOOLS

FIXEDSplit Grid Dialog Fix: Fixed split grid dialog getting cut out on small screens

IMPROVEDCreative Director Auto Scroll: Auto scrolling disabled in Creative Director after a grid split operation for better UX

FIXEDCreative Director Mobile Media Library: Fixed mobile media library issues in Creative Director

Universal Split Grid: Split grid option now available for any individual image anywhere in the app

NEWQwen Max Image Models: Added Qwen max image models in text to image and image editing

FIXEDCreative Director Cancel Generation: Fixed cancel generation functionality in Creative Director

FIXEDCreative Director Split & Upscale Fix: Fixed duplicated image issue when splitting and upscaling

Creative Director Default Grid: Grid 2x2 is now the default setting in Creative Director

Creative Director Image to Video: Image to video button on an image now moves the image to the container of the prompt

Reference Image Reordering: Ability to drag and swap reference images with each other

Image Reference Row Swap: Now you can swap between order of images in image reference row

Prompt Input Image Editing: Click on image uploaded in images row in prompt input to apply changes to it

VIDEO TOOLS

NEWLTX 2 Audio to Video: New LTX 2 audio to video models added

Pixverse 5.6: Pixverse model updated to version 5.6

Video Editor Optimizations: Performance optimizations in the video editor

NEWDubbing Model: New dubbing model added in lipsync-video models

AUDIO TOOLS

FIXEDMusic Generation UX/UI Fixes: Fixed various bugs in music generation UX/UI

NEWMusic Generation Presets: New music generation presets system with 52 new presets

FIXEDText to Speech Emotion Tags: Fixed text to speech emotion tags functionality

CHAT & AI ASSISTANTS

Chat URL Analysis: Fixes in chat tools for URLs and landing pages analysis

FIXEDFile Creation in Chat: Bug fix in files creation inside the chat - simply say create docx or create srt and it will generate the file

IMPROVEDChat Queue System: Queue system implemented in chat tools for better message handling

Message Deletion: Deletion of messages in the middle of conversation now works in help widget and chat tools

MAJORChat Image Creation Redesign: Image creation dialog redesigned with improvements in chat tools

GENERAL

FIXEDStability Fixes: Stability fixes across all the app

Crypto Payments: KOLBO coin cryptocurrency payments now supported

NEWGrok Models Update: All new Grok models updates (text to image, image editing, video to video and image to video)

Liquid Glass UI: Liquid glass UX/UI standard updates across the whole app

NEWMedia Library Categories: New categories in media library - motion graphics and video editor exports

Help Widget Queue: Queue system in the help widget - send multiple messages without waiting for each to finish

RTL & Translation Fixes: Full translations fixes and RTL fixes in languages like Hebrew and Arabic

IMPROVEDPresets Algorithm: Better presets algorithms for improved recommendations

NEWWebsites Media Page: New websites media page to see all your shared artifacts pages

January 25, 2026

ORGANIZATION & WORKFLOW

SSO Login for Organizations: Enterprise single sign-on (SSO) authentication now available for organizational accounts, enabling seamless and secure team access

NEWOrganization Registration: New streamlined registration process for organizations to set up their team accounts

Organizations Custom Assets: Organizations can now create and manage Visual DNA, moodboards, and custom agents for their teams

Global Custom Assets: System-wide support for global custom agents, moodboards, and Visual DNA accessible across all users

FIXEDAuthentication Logout Fix: Resolved bug that incorrectly logged out super admins when other organization members logged out

AI MODELS & FEATURES

NEWMotion Graphics Models: New motion graphics models now live in text-to-video generation for enhanced creative capabilities

Prompt Moderation System: Advanced content moderation and censorship controls with 4 configurable levels of prompt filtering

AI Auto-Enhance Toggle: Users can now disable AI auto-prompt enhancement in various tools, allowing raw input without AI modifications

IMAGE TOOLS

FIXEDNano Banana 4K Bug Fix: Fixed critical bug in image editing tool that incorrectly converted files to JPEG format

USER INTERFACE

IMPROVEDMenu Clarity Improvements: Enhanced UX/UI across all menus for clearer navigation and better user experience

IMPROVEDVisual DNA & Moodboard UX: Improved interface design and usability for Visual DNA and moodboard management

Media Library Delete Controls: Delete buttons in video and image detail sidebars now only visible within media library context, with fully functional deletion

FIXEDFavorites Media Library Fix: Resolved display issue where checkboxes appeared grouped for audio and video files instead of showing individual file controls

GENERAL

NEWWhite Label Lite Version: New streamlined lite version of Kolbo.AI designed specifically for white label organizational deployments

User Logo Removal: Removed user-customizable logo option to maintain consistent branding experience

Credit Transfers: Users can now transfer credits to each other, enabling collaborative credit sharing and team resource management

January 18, 2026

IMAGE TOOLS

IMPROVEDEnhanced Mention System: Dynamic mention system with thumbnail images and Visual DNA references for better context

IMPROVEDVisual DNA Enhancements: Now supports character details like age, hair color, and more with improved algorithm for better character consistency

Image Editing & Creative Director: Integrated mention system with image analysis during generation to provide full context of what's actually in each image in the user's prompt

NEWEnhancer Tool: New properties and upgraded to v4 model for improved image enhancement quality

IMPROVEDCreative Director Optimizations: Bug fixes and optimizations for Visual DNA usage with improved descriptions and context

VIDEO TOOLS

FIXEDVideo Tools Bug Fix: Fixed issue where generations failed when users selected media images larger than 10MB

NEWVeo 3.1 Enhancements: New ingredients (elements) feature with 9:16 aspect ratio support and upgraded model performance

FIXEDVeo 3.1 Extend: Bug fixes and stability improvements for more reliable video extension

AI MODELS

NEWFlux Klein: New Flux Klein model added for advanced image generation

NEWImagineArt: New ImagineArt model added to the platform

BUG FIXES

FIXEDVarious Bug Fixes: Stability improvements and bug fixes across multiple tools and features

IMPROVEDImproved Descriptions: Enhanced descriptions and context throughout the platform for better clarity and understanding

January 15, 2026

VIDEO TOOLS

NEWByteDance Video Upscale Model: New high-performance video upscaling model with enhanced quality

IMPROVEDVideo Upscale Speed Improvements: Significantly faster video upscaling processing times

IMPROVEDVideo Upscale Bug Fixes: Resolved stability issues and improved reliability

IMAGE TOOLS

IMPROVEDCreative Director Optimizations: Enhanced versatility for more creative variations and outputs

IMPROVEDCreative Director Text Rendering: Dramatically improved text rendering quality and accuracy

IMPROVEDVisual DNA Fixes: Resolved issues in Creative Director Visual DNA for better character consistency

NEWZoom Out Feature: New zoom out capability for Reframe tool - uniformly expand any image

NEWMultiple Aspect Ratio Variations: New zoom out model supports many more reframe aspect ratios

ORGANIZATION & WORKFLOW

Media Library Custom Folders: Create and organize media in custom folders

Move Items Between Folders: Seamlessly move items between custom folders in media library

Move Items Between Projects: Transfer media items across different projects

IMPROVEDMove Sessions Between Projects: Relocate any session to any project for better organization

CREATORS HUB

NEWSocial Link Field: New social link field for linking creator profiles

IMPROVEDFeature Requests Section: Enhanced feature request tracking and submission system

IMPROVEDBug Reports Section: Improved bug reporting workflow for better issue tracking

DESKTOP APP

Format Factory Integration: Full format conversion toolkit for any media type

Universal Media Converter: Convert any image, video, or audio file to any format

Trim Tool: Built-in trimming for video and audio files directly in desktop app

AI MODELS

Venice LLM Updates: Improvements and optimizations to uncensored LLM capabilities

USER INTERFACE

NEWHelp Widget Manual Model Selection: New manual model selection capability in the help widget for better control

BUG FIXES

FIXEDMagic Edit Bug Fixes: Various stability improvements and bug fixes in Magic Edit tool

Speed Optimizations: Performance improvements across video upscaling and editing features

January 13, 2026

IMAGE TOOLS

NEWQwen Text-to-Image: New high-quality text-to-image model with exceptional detail

Grok Image Editing: Advanced image editing model for professional-grade modifications

NEWKling O1 Image Editing: New cutting-edge image editing model

Creative Director Visual DNA: Full support for Visual DNA in creative sets

Creative Director Moodboards: Integrate moodboards into your creative workflow

Creative Director Presets: Apply presets directly in Creative Director sessions

IMPROVEDImage Reference Algorithm: Fixed and optimized style reference interpretation

NEWAngles Variations: New model and algorithm for improved multi-angle generation

Magic Canvas Batch Download: Select and download multiple images at once

VIDEO TOOLS

NEWLTX-2 Model: New video model with custom aspect ratio support

IMPROVEDVideo-to-Video Optimizations: Enhanced stability and algorithm improvements

NEWFFmpeg Server: New dedicated server reduces load and improves video-to-video performance

AUDIO TOOLS

Favorites System: Add favorites to text-to-speech, audio studio, and music generations

IMPROVEDSpeech-to-Text Fixes: Enhanced reliability and optimizations

CHAT & AI ASSISTANTS

MAJORChat Image Creation Redesign: Complete overhaul of prompt input and options

Moodboards in Chat: Full moodboard support in chat tool image creation

Presets in Chat: Apply presets directly when creating images in chat

Visual DNA in Chat: Integrate Visual DNA into chat image generation

All Aspect Ratios: Support for all aspect ratios in chat image creation

Realtime Transcription: Extended limits in chat tools, unlimited duration in AI Docs

GENERAL

NEWDesktop App Release: New version with performance improvements and bug fixes

FIXEDAdobe Plugin: Bug fixes and optimizations for smoother workflow

Video Editor: Exports feature officially out of beta

NEWCustom Icons: New custom icons implemented across entire app

NEWCreators Hub: Complete redesign with new features and improved UX

NEWFeature Requests Section: New section where users can request features and earn credits

FIXEDBug Report Section: Report bugs and earn credits for helping improve the platform

FIXEDCreative Director Sessions: Fixed bug preventing sessions from loading properly

FIXEDShared Projects Media: Fixed bug - media now properly shared between users

Custom Logo: User-uploaded logos now reflect across entire app

FIXEDVarious Bug Fixes: Stability improvements and minor bug fixes across the platform

January 4, 2026

IMAGE TOOLS

Wan 2.6: Uncensored high-quality text-to-image and image editing model

NEWFlux 2 Flex: New text-to-image and image editing model

IMPROVEDCreative Director Algorithm: Smarter creative set generation with better results

Creative Director Image References: Reference specific images in manual mode

IMPROVEDCreative Director Parallel Generations: Generate multiple assets simultaneously for faster workflows

NEWCreative Director Enhanced Editing: Edit directly in session without opening new pages

IMPROVEDCreative Director Collapsed Prompt: Improved stability and reliability

Multi-Model Reference Support: Use reference images across supported models (Nano Banana, GPT, and more) with intelligent model switching

IMPROVEDEnhanced Upload Compatibility: Support for various image formats and resolutions in Creative Director

IMPROVEDAdvanced Style Analysis: Improved algorithm for style reference interpretation in text-to-image

Interactive Image Popup: Draggable popup view when clicking images in editing workflows

Video Frame Extraction: Drag and drop videos to instantly extract frames as images

Profile Image Cropper: Now fully stable and functional

VIDEO TOOLS

Kandinsky 5: Advanced video models in text-to-video and image-to-video

HeyGen Lipsync: Professional-grade lip-sync technology integration

Universal Aspect Ratio Control: Auto-detection and manual selection across all image-to-video modes including Creative Director

Smart Model Selection: Automatic model selection in Elements and First Last Frame tools

Face Swap in Videos: Transform characters in video content

Webcam Recording: Direct webcam capture in video-to-video prompt input

Real-time Progress Tracking: Accurate percentage updates for lipsync and video-to-video generations

Lipsync Video Support: Direct video attachment for audio reference without conversion

IMPROVEDVideo Creation Stability: Enhanced performance and reliability

Image-to-Video Controller: Prompt controller fixes and optimizations

FIXEDElements Mode: Aspect ratio issues resolved

AUDIO TOOLS

MAJORAudio Studio: Complete Kits AI integration with voice isolation, music stem creation, and voice transformation (100+ voice options)

MiniMax Voices: Multi-language voice support including Hebrew

MiniMax Voice Cloning: Clone voices and create custom voices from text prompts

Eleven Labs Scribe: Next-generation transcription engine for superior accuracy

Speech-to-Speech Cloning: Advanced voice cloning in Audio Studio

Custom Voice Library: Save and store custom voices in voice cloning tool

Video Audio Extraction: Attach video files to any audio component for automatic audio extraction

IMPROVEDText-to-Speech Bug Fixes: Resolved various bugs for improved reliability

CHAT & AI ASSISTANTS

Native Web Search in Chat: LLMs automatically perform web searches when detecting search intent

IMPROVEDEnhanced Perplexity: Improved performance and more accurate web search results

Mention System: Tag and reference assets with @image1, @image2 syntax across all compatible tools

Real-time Editing: Live collaborative editing in AI Docs, Projects, and Chat Tools

Live Transcription Mode: Real-time microphone transcription in AI Docs

Edit Tracking in AI Boards: Visual banners showing edits made to images

Message Input Redesign: Expandable message view with refined background styling

Chat Archiving: Archive conversations directly from media library

IMPROVEDMemory System Upgrade: Smarter algorithms with enhanced performance

FIXEDLarge File Support in Chat: Fixed bugs with large image files in chat sessions

FIXEDVideo Analysis in Chat: Resolved bugs across all LLM chat models

FIXEDDialog Improvements: Fixed rename and deletion dialogs in sessions sidebar

FIXEDChat Message Deletion: Fixed bugs in message deletion

GENERAL

Team Credits: Collaborate with your team using shared account credits for seamless teamwork

Trash Bin: Safely recover deleted assets with 30-day storage protection

Creators Hub: Marketplace platform for AI creators to offer services to clients [Coming Soon]

FIXEDMedia Library Stability: Various optimizations across media management

FIXEDDesktop App Optimization: Speed, memory, and stability improvements

Mobile UX/UI Improvements: Comprehensive mobile interface enhancements

IMPROVEDProject Creation: Enhanced algorithm with better description generation

FIXEDEmail Notifications: Resolved sending issues

FIXEDImage Editor Zoom: Stability improvements for unexpected zoom behavior

December 29, 2025

IMAGE TOOLS

Layers feature: Photoshop-quality separation with one click - automatically decompose images into up to 10 separate transparent PNG layers

December 28, 2025

VIDEO TOOLS

Kling O1: Video-to-Video editing with up to 2-minute outputs at 1080p/30fps, character swap, style transfer, VFX effects, multi-asset smart tagging (up to 4 reference images), native audio sync, and professional motion preservation

December 26, 2025

IMAGE TOOLS

IMPROVEDCreative Director: Improved AI logic for better creative set generation, enhanced prompt understanding for more accurate results, faster processing and better quality control

IMPROVEDImage Generation: Upgraded image creation with smarter model selection, better handling of complex prompts, improved quality and consistency across all models

IMPROVEDTraining Lab: Enhanced custom model training capabilities, better integration with Flux and other models, improved training results and faster processing

NEWVisual DNA: New feature to save and reuse your unique visual styles, maintain consistent characters across projects, apply your signature look to new creations instantly

NEWMoodboards & References: Improved moodboard manager for easier organization, smarter reference image analysis, new and improved style presets

VIDEO TOOLS

IMPROVEDVideo Generation: Major upgrades to all video models (Kling, Hailuo, Seedance, Sora 2) with better video quality and generation speed, improved handling of first/last frame animations, enhanced video-to-video transformations

NEWElements & Motion: New element combination features, better motion transfer capabilities, improved multi-element animation handling

IMPROVEDLipsync & Dubbing: Enhanced voice synchronization, better lip movement accuracy, improved multi-language support

AUDIO TOOLS

IMPROVEDText to Speech: Completely redesigned voice selection interface with easier voice browsing and preview, better voice cloning capabilities, improved audio quality across all models

IMPROVEDMusic Generation: Enhanced Suno integration with better prompt understanding for music creation

CHAT & AI ASSISTANTS

IMPROVEDChat & AI Assistants: Better visualization of generated code and documents, smoother switching between AI models, improved understanding of your project needs

GENERAL

NEWMobile Experience: Completely new mobile interface with bottom sheet navigation, better touch controls for phone and tablet users, optimized layouts for all screen sizes

NEWTrash & Recovery: New feature to safely delete and recover your media, restore accidentally deleted projects, better media management and storage control

IMPROVEDCredits & Billing: Smart credit management and tracking, better credit reservations during generation, enhanced subscription handling, fixed duplicate credit refresh issues

IMPROVEDSafety & Content: Enhanced content safety with improved NSFW detection, smarter content moderation for safer results, better error messages when content doesn't meet guidelines

NEWCampaigns & Promotions: New campaign system for in-app promotions, personalized discounts and upgrade opportunities, enhanced KOLBO30 promo code system

IMPROVEDMedia Management: Better organization with improved media hub, enhanced media upload and import system, better connection between projects and media, faster media loading and preview

NEWUser Interface: New chip design for easier model browsing, quick and intuitive aspect ratio selection, better progress indicators during generation, improved dark mode consistency

IMPROVEDPerformance & Stability: Faster generation across all creative tools, clearer error messages and recovery options, fixed bugs and stability issues, automatic model switching when needed

December 26, 2025

IMAGE TOOLS

Higgsfield Suite: Three powerhouse models - Soul (image generation), Popcorn (image editing), and DoP (Director of Photography with 100+ cinematic presets)

December 25, 2025

VIDEO TOOLS

Seednace 1.5: Text-to-Video and Image-to-Video model with native sound generation

December 23, 2025

VIDEO TOOLS

Cinema Mode: Generate up to 8 cinematic shots with full narrative consistency, character lock, and parallel rendering

Kling Video 2.6 Motion Control: Up to 30 seconds of one-take motion control with synchronized motion, expression & lip sync

December 22, 2025

IMAGE TOOLS

NEWFlux 2 Models: New Flux 2 Flash and Flux 2 Turbo models added in text-to-image and image editing

NEWHiggsfield Soul: Added in text-to-image with reference image support and Higgsfield presets

NEWHiggsfield Popcorn: Added image model by Higgsfield in image editing tool

NEWLayers Compositing: Added compositing model to image editing

VIDEO TOOLS

Cinema Mode in Creative Director: Special cinema mode with manual and auto-pilot modes, support for up to 8 scene generations, able to generate concurrent parallel image-to-video generations or separate scenes with first and last frame control

Wan 2.6 Video-to-Video: Support for 3 videos and elements in video-to-video tool

Wan 2.6 Audio Input: Support for audio input in image-to-video and text-to-video

NEWHiggsfield Video: Added Higgsfield presets and video model in image-to-video

Video Tools Fixes: Users can now send video generations with presets without prompts, also in first-last frame

FIXEDSora 2 Storyboard: Bug fixes in Sora 2 storyboard scene builder UX/UI

CHAT & AI ASSISTANTS

PDF Watermark Removal: Removed watermark in AI Docs PDF export

GENERAL

NEWDesktop App: Added individual refresh options

FIXEDPlugin Fixes: Fixed bugs in in and out auto-cut + link keeping of clips in the plugin

FIXEDStability Fixes: Overall stability fixes in plugin and desktop app

Affiliate Program: URL bugs fixes

IMPROVEDServer Upgrades: Better speed and stability improvements

December 19, 2025

GENERAL

FIXEDServer maintenance and upgrade to improve performance and stability

December 18, 2025

IMAGE TOOLS

FIXEDCreative Director: Bugs fixed for manual mode + aspect ratio support for image creation in manual mode

Creative Director: Auto-Prompt / Enhance Prompt feature

Canvas Tools: Debugging and optimizations - more stable and shows result images properly

MAJORCanvas Tool: Layout redesigned completely

FIXEDMidjourney: Stability optimizations

Midjourney: Removed native image reference capabilities

FIXED3D Multi-Image: Bug fixes

IMPROVEDImage Editing: Visual DNA shown in better design

IMPROVEDImage Editing: Overlay animation when user hovers on image references fixed with improved UX/UI

NEWOpenAI Image 1.5: New model support added

VIDEO TOOLS

NEWWan 2.6: Added multi-shot + sound text-to-video and image-to-video model

Video to Video: Explanation previews of images and videos explaining each model and what it's doing

AUDIO TOOLS

NEWText to Speech: 100+ new voices from Google and OpenAI with multi-language support

IMPROVEDText to Speech: Massive UX/UI improvements and optimizations for faster and better performance

Text to Sound: Now allows concurrent generations - users don't need to wait until generation finishes to send the next one

Suno Music Generation: WAV conversion, multi-track generation of existing song up to 12 tracks separated, music video lyrics generation from existing song, and extend existing track created by Suno with prompt and guidance

CHAT & AI ASSISTANTS

NEWChat Models: GPT 5.2 and DeepSeek new models added, plus optimizations to all chat tools

Chat Tools: Users can now select text while a response is streaming

NEWClaude Opus 4.5: Added to chat tools

FIXEDGemini Flash 2.5: Markdown bugs fixed

Gemini Pro 3: Fixing and stabilizations

GENERAL

Cancel Any Generation: Cancellation option for all generation types (image, videos, chat). Users can cancel generations and get credits back without affecting them, even if made by mistake

FIXEDGlobal Progress Panel: Bugs fixed - now shows proper images or video thumbnails and directs users to their session generations

IMPROVEDProject Creation: Process simplified with better UX/UI

Fallback Models: If one API provider fails, another provider tries to generate with same model - no extra cost for users

Media Library: Items now show the prompt used for making them

NEWNew Sidebar: Details, prompt, and quick action buttons (video creation, upscale, favorite, edit) for media library images

FIXEDStability Fixes: Tons of stability fixes and reengineering of canvas tool

Media Library Navigation: Users can now navigate to the session their creation was made from

IMPROVEDProjects Creation: Simplified and made better

December 7, 2025

VIDEO TOOLS

Kling 2.6: Native Audio-Visual text/image to video with built-in dialogue, ambient sound, and effects

Kling O1: Precision Cinematic Control with frame-level precision and first/last frame locking

Kling Avatar v2: Next-Gen Character Creation with talking-head avatars, lifelike expressions, and automatic lip-sync

Seedance 4.5v: Stylized Motion Mastery with advanced motion synthesis and rich stylistic control

November 26, 2025

IMAGE TOOLS

FLUX.2: Studio-grade photorealism with perfect text rendering, multi-reference consistency (up to 10 images), lightning-fast generation (under 10 seconds), up to 4MP resolution, and advanced image editing capabilities

November 20, 2025

IMAGE TOOLS

Nano Banana 2 Pro: AI images with perfect text rendering, 4K resolution (2K/4K export), Pro controls (lighting, camera angles, color grading), multi-step auto-refinement, cultural context awareness

November 19, 2025

IMAGE TOOLS

Camera Angles: Turn 1 image into 15 different angles with no 3D modeling required

Topaz Upscale: Professional-grade AI upscaling with 8 specialized models, up to 4x/8K resolution

Real-time image previews: Images appear instantly as ready (no more batch waiting)

VIDEO TOOLS

Creative Director → Videos: Animate up to 8 images in one action for B-rolls, product showcases, and trailers

IMPROVEDKling Turbo 2.5 Pro: Upgraded First Last Frame with faster, higher-quality video generation

IMPROVEDVeo 3 Video Extend: Major stability improvements with fixed socket management, stable progress tracking, enhanced error handling

AUDIO TOOLS

IMPROVEDMusic Generator Overhaul: Extend tracks seamlessly, vocal gender control, audio influence settings, smart lyrics transcription, socket stability fixes, immediate save, faster real-time updates

CHAT & AI ASSISTANTS

GPT-5.1 (OpenAI) now in Chat Tool

Gemini 3 Pro (Google) now in Chat Tool

Switch models without losing context

GENERAL

Black Friday: 30% off all plans with code BLACK30

November 16, 2025

IMAGE TOOLS

Camera Angles: Transform a single image into up to 15 angles instantly with one click, real-time 3D preview, simple controls (rotate, tilt, zoom), no 3D modeling needed

November 11, 2025

GENERAL

NEWNew Media Library: Centralized hub for all AI-generated content with smart organization, advanced search and filtering, bulk operations, custom collections, and quick access to all your media

November 7, 2025

GENERAL

Black Friday Deal: 30% off coupon (code: BLACK30) valid through all November for any one-time purchase

October 24, 2025

IMAGE TOOLS

MAJORTraining Lab Algorithm Major Upgrade: Automatically generates hidden 'character prompt' behind the scenes for consistent character look

VIDEO TOOLS

LTX V2 Models: Latest models capable of stunning 4K resolution and up to 50 FPS for high-quality video generations

GENERAL

IMPROVEDStability & Design Improvements: Rock-solid stability and sleeker interface optimized for smaller screens (mobile/tablets)

October 23, 2025

IMAGE TOOLS

FIXEDImage to Image Tool Fix: Issues with loading sessions containing many images resolved

CHAT & AI ASSISTANTS

IMPROVEDPerformance Boost: Significant optimizations made, especially in Chat Tools, making the platform much faster and more responsive

Queued Messages: Chat now has a message queuing system to maintain proper context

GENERAL

MAJORPrompt Copilot Widget: Redesigned help widget that's dynamic and provides tool-specific guidance

Hide Input/Prompt Box: Hide the main input/prompt area in all tools for more viewing space on smaller screens

NEWNew Apps Section: Added to main menu as hub for tools like AI Board, AI Docs, and Video Editor

FIXEDGeneral Stability & Bug Fixes: Various bugs fixed including page refresh issue when switching tool categories

October 19, 2025

IMAGE TOOLS

NEWNew Image Model: Brand-new image model with exceptional quality

Extend Midjourney Videos: Now available directly within Kolbo.AI

VIDEO TOOLS

Sora 2: Now lets you create videos up to 15 seconds long

Sora 2 Storyboard Feature: Craft epic 25-second stories

Veo 3: Massive boost with video extension & presets for first-last frame and elements

GENERAL

Credit Cost Preview: See the credit cost of your generations before you generate by number of images or duration of video, calculating also multi-model generations

Veo 3 Price Reduction: Slashed prices on veo 3 features

October 16, 2025

VIDEO TOOLS

MAJORWAN Alpha Text-to-Video Model: Complete with parameters and usage examples

NEWAdvanced Video-to-Video Models: Added Animate Replace (swap character in video with character from image) and Animate Move (animate image based on video's motion)

Veo 3.1 Model: Available in First-Last Frame, Elements, Image-to-Video, and Text-to-Video (more credit-intensive than standard Veo 3)

NEWNew Affordable Video Model: Highly affordable model perfect for quickly generating b-roll footage

GENERAL

Delete Any Generation: Delete any unwanted or failed image and video generations across every tool (Text-to-Image, Image Editing, Training Lab, Text-to-Video, Image-to-Video, Lipsync, and more) with soft delete system for safety

Real-time Workspace Updates: Workspace updates in real-time for smooth and fast workflow

NEWMulti-language Support: Added support for Arabic, Spanish, Russian, Chinese, Japanese, Hindi, and more

Affiliate Program: Coming soon - share a link and earn 20% on every purchase

IMPROVEDAnnual Plans: Now available for better value and convenience

FIXEDStability Updates: Pushed many stability updates across the platform

October 7, 2025

GENERAL

MAJORMajor Release: Full Production Ready - Kolbo.AI is officially out of beta and ready for full-scale use

No more storage limits for paid plans

Unlimited projects and agents for everyone - just use your credits

January 22, 2025

IMAGE TOOLS

Visual DNA in Text-to-Image: Visual DNA now works in text-to-image generation

FIXEDVisual DNA Optimizations: Bug fixes and optimizations when using multiple Visual DNA in the same generation

Creative Director Grid Mode: Create up to 32 scenes at one prompt (making 2x2 grid images up to 8 scenes, then users can split and upscale them)

Creative Director Auto Grid Detection: Automatic grid detection in the creative director grid tool

Creative Director Batch Deletion: Deletion now possible for batches in creative director

VIDEO TOOLS

NEWWan 2.6 Fast: New image to video model added

FIXEDVideo-to-Video Bug Fixes: Fixed bugs when uploading videos in video-to-video tool

Video Stitching Reordering: Reordering functionality for stitching of videos operation

ORGANIZATION & WORKFLOW

Organizations Visual DNA: Organizations can now create visual DNAs for their organization

Organizations Custom Agents: Organizations can now create custom agents for their organization

Organizations Moodboards: Organizations can now create moodboards for their organization

Organizations Custom Learning Modules: Organizations now have their own custom learning modules

Global Custom Agents: System support now for global custom agents

Global Visual DNAs: System support now for global visual DNAs

Global Moodboards: System support now for global moodboards

Visual DNA & Moodboards Sharing: Users can now share Visual DNA and moodboards with other users using a shareable link

Project Creation & Editing: Now supports file uploading—images and documents

USER INTERFACE

FIXEDMedia Library Import Fix: Fixed import from library button layout and design issues

Media Library Preview: Click once in media library opens the media item in preview popup

FIXEDRTL Alignments: RTL alignments fixed in all text areas

Menu Realignments: Realignments and tools placements in menus

BUG FIXES

FIXEDVarious Bug Fixes: Stability improvements and bug fixes across multiple tools and features