Docs/Video/Text to Image

Minimax H3· Text to Image

Text to Image

POSThttps://api.sozdai.ai/v1/media/generations

Pricing Model

768P 2,55 ₽/s · 2K 5,1 ₽/s × 4–15s

Execution Flow

Asynchronous submission model. Submitting requests queues background workers.

#Submit a task

Media generation is asynchronous: you get a task id immediately, without waiting.

bash

curl https://api.sozdai.ai/v1/media/generations \
  -H "Authorization: Bearer $CORRY_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "minimax-h3", "prompt": "a cute robot mascot, flat vector logo" }'
# → { "id": "cmus7q0x9000108l3f5g2abcd", "status": "queued" }

#Request parameters

Parameter	Type	Required	Default	Description
`prompt`	string	Yes	—	Text prompt describing the video (max 7000 chars). Required in every mode.
`first_frame_url`	string	No	—	First-frame image URL (image-to-video). JPG/PNG/WEBP/HEIC ≤30MB, 256–5760px, W/H ratio 0.4–2.5.
`last_frame_url`	string	No	—	Last-frame image URL; must be paired with first_frame_url.
`reference_image_urls`	string[]	No	—	Reference images (max 9). Mutually exclusive with first/last frame.
`reference_video_urls`	string[]	No	—	Reference videos (max 3, each 2–15s, total ≤15s; MP4/MOV ≤50MB). Mutually exclusive with first/last frame. Per-second price doubles when reference video is used.
`reference_audio_urls`	string[]	No	—	Reference audio for voice/timbre (max 3, each 2–15s, total ≤15s; WAV/MP3 ≤15MB).
`resolution`	string	No	`768P`	Output resolution; higher costs more per second — see the pricing page. Options: 768P, 2K
`duration`	number	No	`5`	Video duration in seconds, integer 4–15. Billed per second.
`ratio`	string	No	`adaptive`	Aspect ratio. Text-to-video requires a concrete ratio (we default to 16:9 if omitted); image-to-video always follows the input image (adaptive). Options: adaptive, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
`callback_url`	string	No	—	Optional public http(s) URL. We POST the task result here the moment generation finishes (success or failure).

#Poll for the result

Generation runs asynchronously — the submit call returns immediately with a task id. Poll this endpoint with that id every few seconds until status becomes succeeded (result ready) or a terminal error (failed / timeout).

GEThttps://api.sozdai.ai/v1/media/generations/{id}

bash

# take the id from the submit response, then poll every ~3s
curl https://api.sozdai.ai/v1/media/generations/cmus7q0x9000108l3f5g2abcd \
  -H "Authorization: Bearer $CORRY_KEY"
# repeat until "status" is "succeeded" (or "failed" / "timeout")

#Response example

On submit

json

{
  "id": "cmus7q0x9000108l3f5g2abcd",
  "status": "queued",
  "type": "video",
  "model": "minimax-h3"
}

On success (after polling)

json

{
  "id": "cmus7q0x9000108l3f5g2abcd",
  "status": "succeeded",
  "type": "video",
  "model": "minimax-h3",
  "result": { "urls": ["https://cdn.sozdai.ai/video/1782537315223-b49c0c1b771f.mp4"] },
  "cost": "1.6000",
  "currency": "USD"
}

Response fields

Field	Type	Description
`status`	string	Task status: queued / submitted / succeeded / failed / timeout.
`type`	string	Media type: image / video / music.
`model`	string	The model you requested.
`result`	object	Present on success. result.urls is an array of permanent file URLs.
`error`	string	Error message, present when the task failed.
`cost`	string	Amount charged for this task — a USD string (e.g. "0.1200"). See the currency field.
`currency`	string	Currency of cost — always "USD".