Video from Image

Generate video clips from a reference image. Provide a starting frame (and optionally an end frame) along with a motion prompt, and the model will animate the scene.

Caution

Credits are pre-charged when a generation is created. If the generation fails or is stopped, credits are automatically refunded.

Endpoint

POST /v1/vidsheet/{id}/cells/{cell_id}/generate?agent_id={agent_id}

Path parameters

Parameter	Type	Description
id	integer	The sheet ID.
cell_id	integer	The cell ID.

Query parameters

Parameter	Type	Description
agent_id	integer	The agent ID.

---

Models

Model	generation_type	data.model
Kling 2.1	video_from_image	"kling_2_1"
Kling 2.6	video_from_image	"kling_2_6"
Veo 3	video_from_image	"veo_3"
Veo 3.1	video_from_image	"veo_3_1"
Sora 2	video_from_image	"sora_2"
Seedance Lite	video_from_image	"seedance_lite"
Seedance Pro	video_from_image	"seedance_pro"
Seedance Pro 1.5	video_from_image	"seedance_pro_1_5"
Grok	video_from_image	"grok"

---

Request body

Field	Type	Required	Description
generation_type	string	Yes	"video_from_image".
data.image_resource_id	string	Yes	Starting frame image content resource ID.
data.image_tail_resource_id	string	No	Optional end frame image content resource ID.
data.prompt	string	No	Text description of the desired motion or scene. Supports {{variable_name}} syntax.
data.negative_prompt	string	No	What to avoid in the generated video.
data.model	string	No	Model variant. See Models table.
data.aspect_ratio	string	No	"1:1", "9:16", or "16:9". Grok also supports "4:3", "3:4", "3:2", and "2:3".
data.resolution	string	No	Output resolution. Use "1080p" or "720p" when supported by the selected model. If omitted, the API uses the cell default or the closest supported model default.
data.duration	integer	No	Video duration in seconds.
data.native_audio	boolean	No	Request native audio when supported by the selected video model. Defaults to false when omitted.
data.image_var	string	No	Variable name to map the image value from a column.
data.image_tail_var	string	No	Variable name to map the end frame image from a column.
data.prompt_var	string	No	Variable name to map the prompt from a column.
data.negative_prompt_var	string	No	Variable name to map the negative prompt from a column.

---

Response (201)

{
  "generation_id": 9030,
  "status": "pending"
}

Once complete, the generated video is available in output_resources. Poll with GET /v1/generations/{generation_id} to check status.

---

Example

curl -X POST "https://api.gen.pro/v1/vidsheet/101/cells/3200/generate?agent_id=42" \
  -H "X-API-Key: your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "generation_type": "video_from_image",
    "data": {
      "image_resource_id": "image_res_123",
      "prompt": "Camera slowly zooms in while the product rotates, soft studio lighting",
      "negative_prompt": "blurry, distorted, text overlay",
      "model": "kling_2_6",
      "aspect_ratio": "9:16",
      "resolution": "1080p",
      "native_audio": false,
      "duration": 5
    }
  }'

Completed generation

{
  "id": 9030,
  "status": "completed",
  "user_job_type": "video_generation",
  "result": null,
  "failed_reason": null,
  "output_resources": [
    {
      "id": 1800,
      "url": "https://cdn.gen.pro/outputs/vid_img2v_001.mp4",
      "thumbnail_url": "https://cdn.gen.pro/thumbnails/vid_img2v_001.jpg",
      "object_type": "video",
      "type": "output"
    }
  ],
  "execution_cost": 4.0
}

Tip

Use image_tail to create smooth transitions between two keyframes. This is useful for product reveals or before/after sequences.

---

Cell configuration

Tip

Cell configuration is set via PATCH /v1/vidsheet/{id}/cells/{cell_id} with additional_attributes in the request body. These defaults are used when generation is triggered.

Field	Type	Description
model	string	Default model: "kling_2_1", "kling_2_6", "veo_3", "veo_3_1", "sora_2", "seedance_lite", "seedance_pro", "grok", etc.
ratio	string	Default aspect ratio: "1:1", "9:16", "16:9". Grok also supports "4:3", "3:4", "3:2", and "2:3".
resolution	string	Default output resolution for supported video models: "1080p" or "720p".
duration	number	Default duration in seconds.
autoGenerate	boolean	Auto-trigger generation when dependent variables change.