Technical Specifications of runwayml-image-to-video
| Specification | Details |
|---|---|
| Model ID | runwayml-image-to-video |
| Provider | Runway |
| Primary capability | Generates video from a starting image using a text prompt and image-conditioned motion generation. |
| API task type | Asynchronous image-to-video generation task started via Runway’s image-to-video endpoint. |
| Supported underlying models in Runway API | Runway’s image-to-video endpoint accepts multiple video models, including gen4_turbo, gen3a_turbo, veo3.1, veo3.1_fast, and veo3. |
| Required inputs | An input image plus model-specific generation parameters; prompt text is commonly used to describe desired motion and scene behavior. |
| Authentication | Bearer API key in the Authorization header. |
| Versioning header | Runway’s API reference requires the X-Runway-Version header, with the documented value 2024-11-06 for the image-to-video endpoint. |
| Output format | Generated video asset returned through task-based API processing. |
| Typical frame rate | Runway documentation lists 24 fps for Gen-3 Alpha image-to-video outputs. |
| Common output resolution | Runway documents 1280×768 for Gen-3 Alpha and both 1280×768 and 768×1280 for Gen-3 Alpha Turbo, depending on orientation. |
| Image input guidance | Runway recommends avoiding reference images smaller than 640×640 px or larger than 4K for image inputs. |
| Usage pattern on CometAPI | Accessed through CometAPI using the platform model identifier runwayml-image-to-video. |
What is runwayml-image-to-video?
runwayml-image-to-video is CometAPI’s model identifier for accessing Runway’s image-to-video generation capability. In practice, this category of model takes a still image as the visual starting point and transforms it into a generated video, typically guided by prompt text that describes motion, camera behavior, atmosphere, or scene evolution.
Runway positions image-to-video as part of its broader video generation stack. Its documentation for Gen-3 Alpha describes image-to-video as a way to animate a supplied image, with prompting focused especially on movement rather than re-describing everything already visible in the frame.
From an API perspective, Runway exposes image-to-video generation as a task-creation endpoint, which means requests typically start a job first and then require polling or retrieval steps to obtain the final video result after processing completes.
Main features of runwayml-image-to-video
- Image-conditioned video generation: Starts from a still image and generates motion-driven video output rather than creating video from text alone.
- Prompt-guided motion control: Works best when prompts describe motion, camera movement, and scene changes to animate the supplied frame coherently.
- Asynchronous task workflow: Uses a start-generation request followed by result retrieval, which fits production systems that queue and monitor long-running media jobs.
- Multiple underlying model options: Runway’s image-to-video API endpoint supports several backend video models, giving developers flexibility around quality, speed, and cost profiles.
- Production-style API authentication: Uses standard Bearer-token authentication and explicit API version headers for controlled integrations.
- Support for portrait and landscape outputs: Runway documents output sizes for both horizontal and vertical generation modes in supported model families.
- Reference-image workflow compatibility: The API is designed around image references, and Runway provides input-size recommendations to help maintain generation quality.
- Extensible creative pipeline fit: Runway’s broader platform includes related video-generation workflows such as keyframes, expansion, and other generation modes, making image-to-video useful as one stage in a larger creative pipeline.
How to access and integrate runwayml-image-to-video
Step 1: Sign Up for API Key
To get started, sign up on CometAPI and generate your API key from the dashboard. After you have an API key, store it securely and use it in the Authorization header for all requests.
Step 2: Send Requests to runwayml-image-to-video API
Use Runway's official API format via CometAPI. The endpoint is POST /runwayml/v1/image_to_video. Include the X-Runway-Version header.
curl https://api.cometapi.com/runwayml/v1/image_to_video \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_API_KEY" \
-H "X-Runway-Version: 2024-11-06" \
-d '{
"model": "gen4_aleph",
"promptText": "Your prompt here."
}'
Step 3: Retrieve and Verify Results
The API returns a task object with a task ID. Poll GET /runwayml/v1/tasks/{task_id} to check generation status, then retrieve the output URL from the completed task response.