Lovart launches China edition: one agent delivers full brand design kits, reference plugin and 100 pro skills
卡尔的AI沃茨 · wechat · 2026-08-31
Design agent Lovart has launched a domestic (China) edition with phone-number login and a major feature upgrade. The author stress-tested it with several fictional brands.
Full brand kit test: given a fictional lifestyle brand "MOODYGOODS", a single brief covering logo, two original animal mascots, product lines, packaging, pop-up store visuals, six Xiaohongshu posts, a motion poster, a 15-second brand film and a campaign website was delivered within about half an hour, stylistically consistent and roughly 70% production-ready.
Inspiration plugin: a new Chrome extension lets users save reference material from Xiaohongshu, Taobao, Amazon and Instagram directly into Lovart's asset library (images, video, PDF, Word, Markdown, links all work as References); the agent classifies what to keep, borrow, or discard, with human oversight.
Skills: 100 curated professional Skills covering scenarios and styles, plus the ability to save successful flows as personal Skills; one-click PNG/JPG to multi-path SVG export into Figma/Illustrator; outputs include static images, carousels, motion, video, runnable HTML pages, PPT, 5-second loop animations and interactive web features.
The author concludes Lovart is chaining scattered design actions into one flow that increasingly mirrors a human design team — deliverables went from "needs heavy rework" to "70% done, just polish the details".
More from Multimodal
- Seedance 2.5 Generates Photorealistic IGI-Style Tactical FPS Gameplay — SimplyAnnisa · 2026-08-31
- ChatGPT Images Can Now Edit Photos with Prompts, Rivaling Photoshop — TawohAwa · 2026-08-31
- Can MiniMax H3 run on 8GB VRAM and 16GB RAM or is it pointless? — Shanq123 · 2026-08-31
- DeepSeek-V4-Flash-Vision-Exp Open Sourced, Topping Multimodal Benchmarks — 智东西 · 2026-08-31
- Fixing AI video face melt: upload raw audio and quote dialogue in prompt — Fragrant-Cheek-4273 · 2026-08-31
- DeepSeek Releases V4-Flash-Vision-Exp Model — amulyabaral · 2026-08-31