We read every Kling prompt guide so you don't have to.

What Kling actually rewards

We pulled this from Kuaishou's official guidance and what works in production. The short version:

• →Reference-first for consistency.
• →Use clean, centered references.
• →Use explicit motion verbs (melt, ignite, morph).
• →Add preserve-shape continuity notes for transformations.
• →Use multi-shot timestamps.

Common mistakes we fix automatically

• Avoid
  Don't use specific numbers ('5 trees').
• Avoid
  Don't request complex motion in one shot — split into shots.

Frequently asked questions

Which versions of Kling does this support?

We support kling-1.6, kling-2, kling-2.1, kling-2.5-turbo, kling-3, kling-o1. We apply the prompt patterns Kuaishou recommends for each, so the rewrite is tuned to the version you're using.

Is my prompt stored or used for training?

No. Prompts are sent to the rewriter, scored, returned, and discarded. We don't train on them and we don't keep them around.

Do I need to know prompt engineering to use this?

Nope. That's the point. Paste what you have, click Rewrite, get back a version that follows Kuaishou's official guidance.

What makes this different from Kling's own "improve prompt" feature?

Built-in optimizers use the model's own preferences. Ours is built on Kuaishou's official documentation and patterns that consistently produce better results in production. We keep rewrites inside the length window Kling responds best to.