XinYu AI (心宇 AI) is a multi-model AI creative platform for creators. At its core is a creative canvas where you generate images, video, audio and text in one place and produce publish-ready work — no switching between tools.

What can XinYu AI do?

On a single canvas: AI image generation, video generation, storyboarding, audio and text creation; with reference-image character locking, free switching between models, native 4K output, and a full pipeline from draft to finished cut.

Which AI models does XinYu AI support?

XinYu AI integrates leading generative models including the full Seedance 2.0 (video), Kling (video), GPT Image 2 and Nano Banana Pro (image), switchable on one canvas — no need to sign up for each provider separately.

There is a free tier to start creating right away; higher usage is via pay-as-you-go credits or a subscription. See the pricing page for details.

How is XinYu AI different from other AI generators?

Most tools are single-purpose generators. XinYu AI is a full creative pipeline: one canvas connecting generation, references, storyboarding and final output — built for creators who actually publish, not just make one image.

Content creators, short-video and short-drama makers, designers, and teams that need to produce AI images and video at scale and consistently.

Does XinYu AI support Chinese?

Yes. The interface is bilingual (Chinese and English), with full Chinese prompt and creative-workflow support.

Can I use XinYu AI via CLI or other tools?

Yes. XinYu AI offers MCP and CLI access, so you can call its generation from MCP-capable tools like Claude and Cursor, and from the command line.

Is XinYu AI expensive?

XinYu AI is pay-as-you-go with credits that never expire, and image, video and audio share one balance — no separate subscription per media type. Compared with signing up and subscribing to each generator separately, the same amount of work costs less. A free tier lets you start immediately, and subscribers get up to 20% bonus on top-ups.

How does XinYu AI pricing work?

Pay-as-you-go credits (shared across image/video/audio) — pay for what you use, credits don't expire; subscriptions are also available with better top-up value. There's a free tier to try first; see the pricing page.

The canvas agent is now available to everyone

Now open to everyone

Open any canvas and click the desk pet in the bottom-right corner, or press ⌘I (Ctrl I on Windows), and the agent panel appears.

What it does: read what is already on your canvas, create generation nodes on request, set the model and parameters, and wire up reference images. Once the nodes are prepared, you press “Generate all” to actually start them — whether and when anything is spent stays your call.

Two brains are free

Gemini 3.5 Flash
DeepSeek V4 Flash

Both are available without paying. Ten stronger brains (Claude Opus 5 / Sonnet 5, Gemini 3.1 Pro, the GPT-5.6 family, Kimi K3, DeepSeek V4 Pro and more) are on the paid tier and are marked as such in the picker.

It tells you the truth about what it did

While nodes are still generating, the task card reads “Media still generating · N still running”, and only says complete once they all finish.
A plan step that could not be ticked automatically is marked ❓ “not auto-confirmed”, with a note that the work may well have been done — we just could not confirm it automatically. No guessing.
Image models do not all offer the same quality tiers; the agent only picks from the tiers a given model actually supports, so the tier and price on the confirm card are exactly what runs and what gets charged.

Tell us when it is wrong

Every reply has a 👎 and a ⚠️ button. If a run goes sideways, or it does something you did not ask for, press one — that is the only way we can see which specific run went wrong.