Scenario Kling Video
Overview
Kling, Kuaishou's video family on Scenario, ships as specialists, eighteen at authoring time: V3
(Omni plus dedicated T2V and I2V tiers), O1, 2.6, motion control, lipsync, and a talking avatar.
Pick the member whose conditioning matches the job, then treat model_schema_get as the contract:
the same parameter changes shape, default, and legality between members. Kling Video to Audio is
audio output, the scenario-audio skill's domain.
Connection and the core loop: see the scenario skill in this repo; model-agnostic video work: the
scenario-video skill. If a sibling skill named here is missing from your available skills, load it
with skill_get (Scenario MCP) or ask the user to install it
(https://github.com/scenario-labs/skills#install); unattended, proceed from tool schemas and flag
the gap.
Quick reference
Pick by job; discover ids with search (target="models", query="kling", public=true):
| Job |
Members |
Every mode in one model, tier via mode |
V3 Omni (standard 720p, pro 1080p, 4k) |
| Text to video at a fixed tier |
V3 T2V Standard / Pro / 4K, 2.6 T2V Pro |
| Animate a still, optional last frame |
V3 I2V Standard / Pro / 4K, O1 I2V, 2.6 I2V Pro |
| Consistent characters from photos |
O1 Reference Images, V3 I2V elements |
| Edit or restyle footage by prompt |
O1 Video Editing / Reference Video, Omni (videoReferenceType: "base") |
| Motion from a video onto a character image |
Motion Control (V3 Pro / Std, V2.6) |
| Talking head |
AI Avatar (image plus audio), Lipsync (video plus audio or text) |
Schema traps (names from live schemas, caps at authoring time):
- V3 dedicated lines take
prompt or multiPrompt, never both; multiPrompt is an array of
{prompt, duration} shots, and shotType (customize or intelligent) exists only on the V3
T2V members and I2V 4K. Omni keeps prompt required; its multiPrompt is a JSON string of up to
6 shots whose durations sum to duration.
duration is a string enum on most members ("5", not 5); Omni alone takes a number, 3 to 15.
The V3 dedicated lines reach 15 seconds; O1 and 2.6 stop at 10.
- On O1 and V3 I2V an element is a
frontalImage plus up to 4 angle referenceImages: both
required on O1, each optional on V3 I2V, where a video can define the element instead.
- Budgets shrink beside a video: O1 Reference Images takes 7 total (elements plus images), the O1
video members 4, and Omni's
referenceImages drops from 7 to 4 next to referenceVideo. Tags
bind by order (@Element1, @Image1); Omni prompts use <<<image_1>>> and <<<video_1>>>.
- A reference clip runs 3 to 10 seconds on Omni and the O1 video members, 2 to 10 on Lipsync (trim
first, per
scenario-video). Omni's referenceVideo defaults to videoReferenceType: "feature",
lending style and camera to a new clip; "base" edits the clip and ignores duration; 4k mode
refuses a reference video either way.
cfgScale (0 to 1, default 0.5): raise toward 0.8 for storyboard fidelity, drop toward 0.3 to let
the model invent.
- Prompt in director order: scene, subject, action, one camera move, then audio and style; natural
sentences beat tag lists, and complex scenes hold together near 5 seconds, not 15.
Audio is a per-member switch
generateAudio defaults true on the V3 dedicated lines and 2.6 T2V, false on Omni and 2.6 I2V, and
does not exist on O1 (editing members carry source audio via keepAudio). It is refused beside
Omni's reference video and beside lastFrameImage on 2.6 I2V; the 4K members bill per second either
way. Voice output is Chinese and English, other languages auto-translate to English: put dialogue in
quotes in the prompt, lowercase for English speech, uppercase for acronyms. For new speech on
existing footage use Lipsync: an audio file, or text with a voiceId, never both.
Motion control reads two inputs
Motion members require a character image and a driving video; characterOrientation names which
input holds the character and, on V2.6, caps the driving clip at authoring time: 10 seconds for
image, 30 for video. Defaults differ (V3 Std says video, the others image), so set it
explicitly. keepOriginalSound decides whether the source audio survives.
Worked example: two-shot character clip from a still
search with target="models", query="kling image to video", public=true. Prefer the newest
non-deprecated hit, e.g. model_kling-v3-i2v-pro (a live hit at authoring time: re-discover each
session).
model_schema_get with that id: multiPrompt shape, duration values, element caps.
upload_asset the start frame and the character's frontal photo (see the scenario skill).
model_run with that model_id, dry_run=true, and
parameters={"startImage": "asset_s", "multiPrompt": [{"prompt": "Wide shot, @Element1 crosses the plaza, slow dolly-in", "duration": "5"}, {"prompt": "Close-up, @Element1 smiles and says: 'we made it'", "duration": "4"}], "generateAudio": true, "elements": [{"frontalImage": "asset_f"}]}
for the cost estimate; re-estimate after changing durations, tier, or audio.
- Repeat
model_run with wait=false, then jobs_wait with the returned job id, re-called with
pending_job_ids on timeout, never a second model_run.
asset_display the output and review both shots with sound.
Common mistakes
- Passing both
prompt and multiPrompt to a V3 dedicated line: exclusive there (Omni alone keeps
prompt required).
- Numeric durations: outside Omni,
duration is the string "5", not 5.
- Expecting audio beside a last frame (2.6 I2V) or a reference video (Omni): exclusive pairs.
- Carrying one member's caps to another: 15 seconds,
elements, and 4K each exist on one member,
not the next.
- Compound camera moves ("dolly in while orbiting"): one dominant move per shot; split the rest
across
multiPrompt shots.
- Overloaded negative prompts: a short artifact list steers; a long one stiffens motion.
- Prompting Omni
base to extend a clip: it edits in place, and no Kling member extends at
authoring time. Continue from the clip's lastFrame id (asset_get) as the start frame of a new
I2V run, then concatenate, per scenario-video.
- Skipping
dry_run on 4K or avatar runs: at authoring time 4K ran several times Standard and
avatar cost spanned a 100x range. Iterate on a Standard tier; no V3, O1, or 2.6 schema carries a
seed, so a 4K re-run of the keeper is a new take, while upscaling it (scenario-video) keeps the
take.
1---2name: scenario-kling3description: "Use when generating or editing video with Kling models on Scenario via MCP: text-to-video, image-to-video with first and last frames, multi-shot sequences with per-shot prompts, consistent characters via elements and reference images, prompt-based video editing, motion transfer from a driving video (motion control, mocap), lipsync to audio or text, talking avatars, native audio and dialogue, or picking 720p, 1080p, or 4K tiers. Keywords: Kling V3, O1, 2.6, Omni, Kuaishou, T2V, I2V, V2V."4license: MIT5metadata:6 source: https://skillmd.com/skills/scenario-labs/scenario-kling7---89# Scenario Kling Video1011## Overview1213Kling, Kuaishou's video family on Scenario, ships as specialists, eighteen at authoring time: V314(Omni plus dedicated T2V and I2V tiers), O1, 2.6, motion control, lipsync, and a talking avatar.15Pick the member whose conditioning matches the job, then treat `model_schema_get` as the contract:16the same parameter changes shape, default, and legality between members. Kling Video to Audio is17audio output, the `scenario-audio` skill's domain.1819Connection and the core loop: see the `scenario` skill in this repo; model-agnostic video work: the20`scenario-video` skill. If a sibling skill named here is missing from your available skills, load it21with `skill_get` (Scenario MCP) or ask the user to install it22(https://github.com/scenario-labs/skills#install); unattended, proceed from tool schemas and flag23the gap.2425## Quick reference2627Pick by job; discover ids with `search` (`target="models"`, `query="kling"`, `public=true`):2829| Job | Members |30| ------------------------------------------ | ----------------------------------------------------------------------- |31| Every mode in one model, tier via `mode` | V3 Omni (`standard` 720p, `pro` 1080p, `4k`) |32| Text to video at a fixed tier | V3 T2V Standard / Pro / 4K, 2.6 T2V Pro |33| Animate a still, optional last frame | V3 I2V Standard / Pro / 4K, O1 I2V, 2.6 I2V Pro |34| Consistent characters from photos | O1 Reference Images, V3 I2V `elements` |35| Edit or restyle footage by prompt | O1 Video Editing / Reference Video, Omni (`videoReferenceType: "base"`) |36| Motion from a video onto a character image | Motion Control (V3 Pro / Std, V2.6) |37| Talking head | AI Avatar (image plus audio), Lipsync (video plus audio or text) |3839Schema traps (names from live schemas, caps at authoring time):4041- V3 dedicated lines take `prompt` or `multiPrompt`, never both; `multiPrompt` is an array of42 `{prompt, duration}` shots, and `shotType` (`customize` or `intelligent`) exists only on the V343 T2V members and I2V 4K. Omni keeps `prompt` required; its `multiPrompt` is a JSON string of up to44 6 shots whose durations sum to `duration`.45- `duration` is a string enum on most members (`"5"`, not `5`); Omni alone takes a number, 3 to 15.46 The V3 dedicated lines reach 15 seconds; O1 and 2.6 stop at 10.47- On O1 and V3 I2V an element is a `frontalImage` plus up to 4 angle `referenceImages`: both48 required on O1, each optional on V3 I2V, where a `video` can define the element instead.49- Budgets shrink beside a video: O1 Reference Images takes 7 total (elements plus images), the O150 video members 4, and Omni's `referenceImages` drops from 7 to 4 next to `referenceVideo`. Tags51 bind by order (`@Element1`, `@Image1`); Omni prompts use `<<<image_1>>>` and `<<<video_1>>>`.52- A reference clip runs 3 to 10 seconds on Omni and the O1 video members, 2 to 10 on Lipsync (trim53 first, per `scenario-video`). Omni's `referenceVideo` defaults to `videoReferenceType: "feature"`,54 lending style and camera to a new clip; `"base"` edits the clip and ignores `duration`; `4k` mode55 refuses a reference video either way.56- `cfgScale` (0 to 1, default 0.5): raise toward 0.8 for storyboard fidelity, drop toward 0.3 to let57 the model invent.58- Prompt in director order: scene, subject, action, one camera move, then audio and style; natural59 sentences beat tag lists, and complex scenes hold together near 5 seconds, not 15.6061## Audio is a per-member switch6263`generateAudio` defaults true on the V3 dedicated lines and 2.6 T2V, false on Omni and 2.6 I2V, and64does not exist on O1 (editing members carry source audio via `keepAudio`). It is refused beside65Omni's reference video and beside `lastFrameImage` on 2.6 I2V; the 4K members bill per second either66way. Voice output is Chinese and English, other languages auto-translate to English: put dialogue in67quotes in the prompt, lowercase for English speech, uppercase for acronyms. For new speech on68existing footage use Lipsync: an audio file, or `text` with a `voiceId`, never both.6970## Motion control reads two inputs7172Motion members require a character `image` and a driving `video`; `characterOrientation` names which73input holds the character and, on V2.6, caps the driving clip at authoring time: 10 seconds for74`image`, 30 for `video`. Defaults differ (V3 Std says `video`, the others `image`), so set it75explicitly. `keepOriginalSound` decides whether the source audio survives.7677## Worked example: two-shot character clip from a still78791. `search` with `target="models"`, `query="kling image to video"`, `public=true`. Prefer the newest80 non-deprecated hit, e.g. `model_kling-v3-i2v-pro` (a live hit at authoring time: re-discover each81 session).822. `model_schema_get` with that id: multiPrompt shape, duration values, element caps.833. `upload_asset` the start frame and the character's frontal photo (see the `scenario` skill).844. `model_run` with that `model_id`, `dry_run=true`, and85 `parameters={"startImage": "asset_s", "multiPrompt": [{"prompt": "Wide shot, @Element1 crosses the plaza, slow dolly-in", "duration": "5"}, {"prompt": "Close-up, @Element1 smiles and says: 'we made it'", "duration": "4"}], "generateAudio": true, "elements": [{"frontalImage": "asset_f"}]}`86 for the cost estimate; re-estimate after changing durations, tier, or audio.875. Repeat `model_run` with `wait=false`, then `jobs_wait` with the returned job id, re-called with88 `pending_job_ids` on timeout, never a second `model_run`.896. `asset_display` the output and review both shots with sound.9091## Common mistakes9293- Passing both `prompt` and `multiPrompt` to a V3 dedicated line: exclusive there (Omni alone keeps94 `prompt` required).95- Numeric durations: outside Omni, `duration` is the string `"5"`, not `5`.96- Expecting audio beside a last frame (2.6 I2V) or a reference video (Omni): exclusive pairs.97- Carrying one member's caps to another: 15 seconds, `elements`, and 4K each exist on one member,98 not the next.99- Compound camera moves ("dolly in while orbiting"): one dominant move per shot; split the rest100 across `multiPrompt` shots.101- Overloaded negative prompts: a short artifact list steers; a long one stiffens motion.102- Prompting Omni `base` to extend a clip: it edits in place, and no Kling member extends at103 authoring time. Continue from the clip's `lastFrame` id (`asset_get`) as the start frame of a new104 I2V run, then concatenate, per `scenario-video`.105- Skipping `dry_run` on 4K or avatar runs: at authoring time 4K ran several times Standard and106 avatar cost spanned a 100x range. Iterate on a Standard tier; no V3, O1, or 2.6 schema carries a107 seed, so a 4K re-run of the keeper is a new take, while upscaling it (`scenario-video`) keeps the108 take.