MiniMax H3 Insert
MiniMax H3 Insert cuts a new 5 to 13 second scene into an existing clip, then returns to the original footage. A prompt, reference images or reference clips guide the new scene.
video · model h3-max-insert · catalog model-catalog@2026-10-06.v5 · pricing customer-pricing@2026-11-05.v15.Endpoint
POST https://api.roasos.com/v1/video/h3-max-insertSend your API key as a bearer token. The key needs the scope api:write. Every paid request also needs an Idempotency-Key header.
The envelope fields quote_only, quote_id, webhook_url and wait_timeout_seconds behave the same on each model endpoint. Refer to Jobs and Quotes and credits.
Good at
Add a product close-up into a talking-head ad, or swap a weak middle beat for a new one. Send video_url (up to 60 seconds, MP4), start_seconds (where the new scene starts, from 1.625) and resume_seconds (where the source continues, from 2.625). Both must fall inside the source. duration_seconds sets the new scene (5 to 13). Add references (up to 9 images) and video_urls (up to 3 clips, 2 to 15 seconds each and 15 seconds together) to guide it. It renders at 480p or 768p, and matches the colors to the source.
Pricing
The unit is one clip. Pricing key: h3-max-insert.
| Request | Credits |
|---|---|
5s_768p | 4 |
9s_768p | 7 |
13s_768p | 10 |
The credits come from the roasOS pricing table. No other field changes the price. To get the exact price of your request, send it with quote_only: true first.
Modes
edit
Provider endpoints: minimax/h3-max/insert-video.
Request fields
| Field | Type | Required | Default | Rule |
|---|---|---|---|---|
duration_seconds | integer enum | no | 5 | 5, 6, 7, 8, 9, 10, 11, 12, 13 |
prompt | string | no | — | ≤ 10,000 characters |
quote_id | string | no | — | UUID |
quote_only | boolean | no | false | — |
references | string[] | no | — | ≤ 9 items |
resolution | enum | no | 768p | 480p, 768p |
resume_seconds | number | yes | — | 2.625–60 |
seed | integer | no | — | 0–2147483647 |
start_seconds | number | yes | — | 1.625–60 |
video_url | string | yes | — | https URL; ≤ 4,096 characters |
video_urls | string[] | no | — | ≤ 3 items |
wait_timeout_seconds | integer | no | 0 | 0–60 |
webhook_url | string | no | — | https URL; ≤ 4,096 characters |
The schema is strict. A field that is not in this table gets 400 unsupported_option. The field names are the roasOS role names. One role has one name in every family.
Media
This endpoint accepts these media roles:
references— style and appearance referencesvideo_urls— reference videosvideo_url— the source video
Every other role is refused: avatar_id, elements, product_id, product_images, reference_images, image_url, end_image_url, scene_id, scene_image, audio_urls, audio_url, voice_id. A refused role gets 400 unsupported_option.
https://cdn.roasos.com that your workspace owns. Any other URL gets 400 reference_media_not_workspace_owned. Refer to Reference media and uploads to learn how to put a file there.Prompt
Limit 10,000 characters.
- Describe only the new scene and how it joins the footage around it.
- Name the references by what they control: "the product in the first image".
Not exposed
These catalog entries describe omitted inputs, pinned values and source constraints. They do not imply that the provider accepts every listed field.
enable_prompt_expansion— pinned toprovider default (on); never sent.color_match— pinned toprovider default (on); never sent.
Result
The result kind is video. Always present: url.
{
"data": {
"id": "gen_…",
"kind": "video",
"model": "h3-max-insert",
"status": "completed",
"result": {
"url": "https://cdn.roasos.com/workspaces/ws_01/result"
}
}
}The files are immutable objects on https://cdn.roasos.com. A result URL is a valid media input for your next request.
Example
The three steps are always the same: quote, run, poll.
# 1. Quote. Nothing is spent.
curl -X POST https://api.roasos.com/v1/video/h3-max-insert \
-H "Authorization: Bearer roasos_sk_live_..." \
-H "Idempotency-Key: 0f1c9e2a-h3-max-insert-0001" \
-H "Content-Type: application/json" \
-d '{"quote_only":true,"prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera.","resume_seconds":2.625,"start_seconds":1.625,"video_url":"https://cdn.roasos.com/workspaces/ws_01/uploads/video_url.mp4"}'
# 200 { "data": { "kind": "quote", "quote_id": "q_7Hk...", "credits": 4, "expires_at": "...",
# "confirmation": { "model": "h3-max-insert", "example": "5s_768p" } } }
# 2. Run. Same body, same Idempotency-Key, quote_id in place of quote_only.
curl -X POST https://api.roasos.com/v1/video/h3-max-insert \
-H "Authorization: Bearer roasos_sk_live_..." \
-H "Idempotency-Key: 0f1c9e2a-h3-max-insert-0001" \
-H "Content-Type: application/json" \
-d '{"quote_id":"q_7Hk...","prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera.","resume_seconds":2.625,"start_seconds":1.625,"video_url":"https://cdn.roasos.com/workspaces/ws_01/uploads/video_url.mp4"}'
# 201 { "data": { "id": "gen_...", "status": "queued", "credits_reserved": 4 } }
# 3. Poll, then read the result.
curl https://api.roasos.com/v1/jobs/gen_.../status -H "Authorization: Bearer roasos_sk_live_..."
curl https://api.roasos.com/v1/jobs/gen_.../result -H "Authorization: Bearer roasos_sk_live_..."The deprecated path
POST /v1/generations still reaches this model for one minor version. It is deprecated. This is how it maps:
POST /v1/video/h3-max-insert with video_url, start_seconds, resume_seconds, prompt, references, video_urls, duration_seconds (5-13), resolution (480p|768p) and seed.
See also
llms.txtfor this model — the same facts as one plain-text file for an agent.GET https://api.roasos.com/v1/catalog/h3-max-insert— the machine-readable form: enums, defaults, limits and prices. Refer to the API reference.- Jobs — how to poll a job and read its result.
- Provider endpoints:
minimax/h3-max/insert-video.
MiniMax H3 Extend
MiniMax H3 Extend adds 5 to 15 seconds of new footage after an existing clip. The prompt describes what happens next. The output is the whole clip with the new footage, or only the new footage.
MiniMax H3 Styles
MiniMax H3 Styles makes video with sound in one of five fixed looks: VHS tape, 1970s hand-painted toon, low-poly 3D, hand-drawn animation or 16-bit pixel art. It renders at 768p, in any whole length from 5 to 15 seconds.