Prompt Library / Talking-Head / Avatar Setup Shot / for Kandinsky Video

Talking-Head / Avatar Setup Shot Prompt for Kandinsky Video

This is the Kandinsky Video version of the talking-head / avatar setup shot prompt: a locked, centered presenter frame built for avatar and lip-sync workflows: even light, clean sides for captions, no drift. It is formatted the way Kandinsky Video takes prompts best. Free — copy it, replace the bracketed parts, and paste it in.

The prompt

✎ Editable — click in, customize or extend this prompt, then copy. Your edits copy with it.Take this prompt into the Generator and customize it to your own specification →

Reference image — keyframe

Attach a reference image: choose a file, drag & drop it here, or click here and paste. It never leaves your browser.

Attach reference image

Push copies this prompt with a reference line naming your image, and downloads the image file — attach it in the model to complete the pair. This page stages the pair for you; it cannot upload into the model itself (sites cannot reach into each other), so the final attach happens there. The image is downscaled to at most 1024px and stored only in this browser, paired with this prompt. If browser storage is full, it is kept for this visit only.

The deliverable — and how to check it

Deliverable: one 10-second clip, chest-up centered framing, the presenter looking into the lens for the full duration, background clean with clear space on both sides for captions. Acceptance check: freeze any frame — head position and framing must be identical across all of them; any camera drift, reframing or background clutter fails the take.

Series: Biography Film Production — Step 3 of 3 — the interview frame. See its place in the full workflow

Using this prompt with Kandinsky Video

Kandinsky 6.0 Video (Sber / Kandinsky Lab, MIT-licensed open weights) generates video and synchronized 44 kHz audio — including lip-synced speech — in one pass, from text or from a still image as the first frame. Write the shot as prose and describe the soundtrack inside it: dialogue goes between <S> and <E> tags. Clips are 5 seconds on the released checkpoints (Pro 29B and Lite 3B), with a separate super-resolution stage taking them to Full HD. Runs locally via Diffusers/ComfyUI, or through GigaChat.

What this route costs — and how to pay less

Cost structure: open weights (MIT) — Kandinsky 6.0 Video runs on your own hardware, so the marginal cost per generation is zero; the Lite checkpoint runs on modest GPUs. The same prompt is pre-configured in this library for every major model, so you can run it on whichever route fits your budget — comparing the structures before you generate is a free way to save money.

Replace the bracketed placeholders — [PRESENTER DESCRIPTION], [SETTING], [BACKGROUND] — with your own details before pasting.

Other versions and customization