Muse Video Style and Camera Presets: Anime, Fantasy, Short Drama, Product Ads

Instead of describing “anime style” or “whip pan to the next scene” in every prompt, add one field to your Muse request: 7 looks and 13 camera techniques, at the same $0.30 for 10 seconds.

Published: November 4, 2026
Last checked: November 4, 2026
Status: Active

Key facts

Item	Details
Looks	7, including anime, Chinese fantasy CG, Western comic, cyberpunk, product footage, and live-action film, or your own description
Camera	13, including whip pans, cuts on action, short-drama hooks, product-ad close-ups, fantasy speed ramps, and subtle acting
Price	Unchanged: $0.30 for 10 seconds, $1.55 for one continuous 60-second clip
Models	`muse-video` and `muse-video-20s` to `muse-video-60s`
Existing calls	Unchanged; without the new fields the output is the same as before

Anime sample: muse-video, 10 s, 16:9, style_preset 2d_anime plus a custom camera move (request under How to use it)

Recipes by use case

Use case	Send	What you get
Short dramas	`micro_drama_hook`, plus `2d_anime` for animated ones	Conflict or a reversal in the first 3 seconds, ending on a cliffhanger
Product ads	`tvc_product_hero` with `ecommerce_ugc`	Real handling such as opening and pouring, ending on a product close-up
Fantasy and martial-arts fights	`xianxia_speed_ramp` with `3d_xianxia`	Wind-up, burst, and clashes that throw up air and water
Story and dialogue	`emotion_micro_acting` or `over_shoulder_axis`	Emotion through glances and breathing, dialogue that keeps the 180° line
30–60 second clips	Nothing for one continuous shot; add `whip_pan` for scene changes	No visible seams by default, whip-pan transitions when you add it

At 720p and 10 seconds, Muse costs $0.30 per clip and Seedance 2.0 mini about $0.86, based on public prices on October 31; see the video generation API comparison.

Who is affected

Teams making short dramas, product ads, fantasy, or anime content: one look and camera style across a whole batch, with shorter prompts.
Existing integrations: no code changes. Without these fields the output is the same as before.
Other video models: MiniMax H3, Seedance, and Wan are not affected.

How to use it

Two extra lines in your usual request are enough, for example:

"style_preset": "2d_anime",
"skill_preset": "whip_pan"

Here is a full request. It asks for an anime look and spells out the camera move with custom_skill; its output is the sample clip above:

curl -s -X POST "https://api.laozhang.ai/v1/videos" \
  -H "Authorization: Bearer $LAOZHANG_API_KEY" \
  -H "Content-Type: application/json" \
  --data-raw '{
    "model": "muse-video",
    "seconds": "10",
    "aspect_ratio": "16:9",
    "prompt": "A girl with a red scarf runs across a rooftop at sunset as birds fly past",
    "style_preset": "2d_anime",
    "custom_skill": "Start on a close-up of her running feet, then tilt up and pull back into a wide shot of the skyline"
  }'

What each field takes:

style_preset: one look ID.
custom_style: your own look in plain text, up to 2,000 characters; it can go alongside style_preset.
skill_preset: one technique ID, or several as a comma-separated string or an array.
custom_skill: your own camera, shot, or transition rules, up to 2,000 characters; it can go alongside skill_preset.

Available looks and techniques

Looks (style_preset):

cinematic_realistic: live-action film
2d_anime: 2D anime
3d_xianxia: 3D Chinese fantasy CG
american_comic: Western comic
cyberpunk_neon: cyberpunk neon
ecommerce_ugc: product and UGC footage

Camera and transition techniques (skill_preset):

Use	IDs
Transitions between parts	`occlusion_wipe`, `cut_on_action`, `match_cut`, `whip_pan`
Camera technique	`dolly_zoom_oner`, `over_shoulder_axis`, `rack_focus_mood`
Genre pacing	`action_choreography`, `micro_drama_hook`, `xianxia_speed_ramp`, `tvc_product_hero`
Performance and continuity	`emotion_micro_acting`, `spatial_blocking_continuity`

What each ID does is described in Set a visual style and camera moves in the Muse guide.

FAQ

Will my output change if I don’t send these fields?

No. Without them no look and no transitions are added, and a 20–60 second clip still runs as one continuous shot with no visible seams. Sending none does the same.

Do these fields cost extra?

No. Muse is priced per video, $0.30 for 10 seconds up to $1.55 for 60 seconds, whether or not you set a look or camera technique. Full pricing is in the Muse guide.

Can I combine several techniques?

Yes, as "whip_pan,cut_on_action" or as an array. none can’t be combined with other techniques. A wrong ID returns HTTP 400 at submission, so no task is created and nothing is charged.

Why does the frame go dark with occlusion_wipe?

That is how it works: at each seam a foreground object such as a tree trunk or door frame covers the whole frame before the next part slides in, which darkens the picture for about 1–2 seconds. Leave it out for a continuous shot, or use whip_pan or cut_on_action for a clear cut without the dark frames.

Meta Muse Video API guide: every field, error, and example.
Video generation API comparison: price, length, and features across video models.
Muse 40–60 second clips (November 2, 2026)
Models and pricing
LaoZhang API announcements

​Key facts

​Recipes by use case

​Who is affected

​How to use it

​Available looks and techniques

​FAQ

​Will my output change if I don’t send these fields?

​Do these fields cost extra?

​Can I combine several techniques?

​Why does the frame go dark with occlusion_wipe?

​Sources and related links