Radar / AI media tools / Canva Craft
Canva Craft: Canva's first in-house image model plans the design before drawing it
Canva launched Craft, its first in-house text-to-image model. It plans a composition from the prompt - layout, scale, hierarchy, on-image text - before rendering, targeting the legible text and precise placement that trip up general image generators.
Why it matters
Plan-then-render architecture trained on millions of human-made design templates rather than photographs, so posters and infographics keep readable text; outputs up to 2K resolution in 30+ aspect ratios, live in Dream Lab and via API.
What you could build with it
An indie developer building marketing tools could offer one-click ad variants for small businesses by generating layout-accurate posters through Craft, then letting owners swap text and branding without re-rendering the whole design.
Does it hold up?
Usable today inside Canva's Dream Lab and via the Connect API (model craft_v1_2k), but each generation takes roughly 50-60 seconds and runs on AI credits, so it fits design work better than bulk pipelines.
Built with Canva Craft
- You Can Now Generate AI Images Using Canva, Here Is HowGizbot · Hands-on how-to: Craft is reachable from Canva's Dream Lab tab, takes prompts describing subject, background, colours and style, and supports targeted edits to one object without changing the rest.
- Canva launches its 1st image generation modelNewsBytes · Launch coverage noting Craft was co-developed with Leonardo.ai researchers, with faster Craftfast and Craft-ultrafast variants planned, plus a future MCP integration.
- Canva launches Craft, its first image model that plans before drawingalextech.ai · Coverage of the two-stage architecture and Canva's claim of fourth place on the DesignArena crowdsourced image benchmark, behind three OpenAI GPT-Image variants.
Learn more
First spotted on article: source.
More AI media tools
VoiceStudio — open-source, fully-local ElevenLabs alternativeOpen-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation…media · JEV 0.73OpenMontage: open-source agentic video production systemTurns an AI coding assistant into a full video studio: 12 pipelines and 100+ tools for scripting, footage, narration…media · JEV 0.73LingBot-Map: streaming 3D reconstruction foundation model (ECCV 2026 best-paper candidate)Feed-forward 3D foundation model that reconstructs scenes from streaming video at ~20 FPS with drift correction over…media · JEV 0.7HyperFramesHeyGen's open-source framework that turns HTML/CSS/GSAP into deterministic MP4 video via 21 agent skills and a router…media · JEV 0.69ArtCraft pivots to seven open-source Adobe clones built with Claude Opus 5.5Brandon Thomas pivoted his AI art IDE into a seven-app open-source creative suite…media · JEV 0.67SolidCraft: open-source parametric CAD in Rust that agents can drive over MCPSolidCraft is an open-source parametric 3D CAD application written in pure Rust (desktop, browser via WebAssembly…media · JEV 0.67
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.