Why We Rebuilt Everything
Studio 1.0 was a collection of tools. It worked, but each editor felt separate — different UI patterns, different AI integrations, different mental models. When we sat down to plan updates, we realized the problem wasn't features. It was architecture.
Studio 2.0 is a rebuild from scratch. Every editor shares a unified data model. Every AI integration goes through a common pipeline. The result is a creative suite that actually feels like one product.
Nine Editors
Canvas Editor
The primary workspace for image creation and editing. Multi-layer support with blend modes, selection tools (rectangle, ellipse, lasso, magic wand, pen), transform operations, and a full filter pipeline. The canvas handles raster and vector content simultaneously.
AI tools are integrated directly into the canvas workflow — inpaint a selection, outpaint to extend a composition, remove backgrounds, upscale regions, or generate variations of selected elements. No export-to-another-tool roundtrip.
Video Editor
Timeline-based editing with multi-track video and audio. Trim, split, merge, speed adjustment, and crossfade transitions. A properties panel controls brightness, contrast, saturation, and audio levels per clip.
AI capabilities include text-to-video generation, video-to-video style transfer, automatic captioning, scene detection, and keyframe interpolation for smooth slow motion.
Audio Editor
Waveform editing with cut, copy, paste, fade, normalize, and noise reduction. The effects rack includes EQ, compression, reverb, delay, chorus, and pitch shift.
AI features: text-to-speech with multiple voices, speech-to-text transcription, music generation from text prompts, voice cloning, audio separation (vocals from instrumentals), and noise removal.
Animation Editor
Keyframe-based animation with easing curves, motion paths, and onion skinning. Supports sprite sheet generation, GIF export, and Lottie JSON output. Frame-by-frame drawing and tweening work together.
Node Editor
A visual programming environment for creative workflows. Connect generator nodes (text-to-image, image-to-image), processing nodes (resize, crop, filter, color adjust), and output nodes (save, export, publish) into reusable pipelines.
Nodes execute in parallel when their inputs are independent. Preview thumbnails update in real time as data flows through the graph. Save node graphs as templates and share them.
Imagine Hub
The unified generation interface. All AI generation models — image, video, audio, 3D — accessible from one panel. Select a model, configure parameters, generate, and send results directly to any editor.
Six standalone generation tools: text-to-image, image-to-image, inpainting, outpainting, upscaling, and background removal. Each tool has its own optimized UI for focused work.
Workflow Editor
A visual pipeline builder for batch creative operations. Chain generation, processing, and export steps. Apply a workflow to a batch of inputs — resize 100 images, apply a consistent style, generate variations, and export to multiple formats in one run.
Batch Editor
Apply operations across multiple files simultaneously. Resize, convert, watermark, rename, compress, and export in bulk. Batch operations can include AI steps — run every image through background removal, apply a consistent style, or generate alt text.
Project Manager
Organize files, collections, and workflows into projects. Asset management with tags, search, and version history. Export entire projects as archives.
43 AI Models Across 7 Providers
Studio integrates with 43 AI models from OpenAI (DALL-E, GPT-4o), Stability AI (Stable Diffusion 3, SDXL), Replicate, Anthropic, Google (Imagen, Gemini), Hugging Face, and Fal.ai.
Each model is tuned for specific tasks. The model picker shows capability badges — which models handle inpainting, which support high resolution, which are fastest for iteration. You can set a default per editor or let the system recommend based on your current task.
100+ AI Tools
Tools span every creative domain:
Image — Generate, edit, enhance, style transfer, colorize, remove objects, expand, segment, depth map, sketch-to-image, face restore
Video — Generate clips, interpolate frames, stabilize, track objects, extract scenes, generate captions, lip sync
Audio — Generate music, synthesize speech, clone voices, transcribe, separate tracks, remove noise, master
Text — Generate descriptions, write alt text, create captions, expand prompts, translate
3D — Generate meshes from images, texture projection, normal map generation
Every tool works non-destructively. Undo, redo, and version history let you experiment freely.
The Pipeline
All AI operations flow through a unified pipeline. A generation request goes through: prompt enhancement (optional), model selection, parameter optimization, generation, post-processing (upscale, format conversion), and delivery to the target editor.
This means any improvement to the pipeline — better prompt engineering, faster inference, smarter post-processing — benefits every editor and every tool simultaneously.
Try Studio 2.0
Available now at studio.poly.inc. Every editor, every tool, no account required to start creating.