Runway Gen-4 Turbo takes an image only: text-to-video on Sume

Runway's API model guide lists gen4_turbo as IMAGE input only, while gen4.5 takes text or image and seedance2_5 takes text, image or video. If you have a prompt and no picture, Gen-4 Turbo is not the model. On Sume, wan-3.0, seedance-2.5, kling-3 and gemini-omni-flash-1.1 all take a text prompt alone.

Input types by model (read 2026-10-04)

Runway's rows come from its model guide. Sume's come from the video models listing and docs.

Text only on Sume

Send the prompt and no frame_images. Add frame_images only when you do have a first frame.

curl -X POST "https://api.sume.com/v1/videos" \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.5",
    "prompt": "A paper plane glides over a city at dusk",
    "duration": 8,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

When you do have an image

Gen-4 Turbo style work, a still brought to life, maps to frame_images with frame_type first_frame. If both frame_images and input_references are present, Sume treats the request as image-to-video.

curl -X POST "https://api.sume.com/v1/videos" \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-3",
    "prompt": "The camera pushes in as steam rises",
    "duration": 5,
    "frame_images": [{
      "type": "image_url",
      "image_url": {"url": "https://example.com/cup.png"},
      "frame_type": "first_frame"
    }]
  }'

Image URLs must be public

Media inputs must be public HTTPS URLs. Upload a local file somewhere reachable first.