Runway Gen-4.5

Runway

Get an API key

Runway's Gen-4.5 text- and image-to-video model.

Modalities: textimagevideo
Price: $1.2/video
Providers: Live

Playground

Preparing playground

Providers

API

POST/api/v1/videosasync jobpoll · GET /api/v1/videos/{id}

runway-gen4.5

# 1 · start the render
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "runway-gen4.5", "prompt": "a paper boat drifting down a neon rain stream", "duration": 8, "aspect_ratio": "16:9" }'

# 2 · poll until it is ready
curl "https://www.ninjachat.ai/api/v1/videos/JOB_ID" \
  -H "Authorization: Bearer $NINJACHAT_API_KEY"

promptmodeldurationaspect_ratioimage_urlreference_imagesreference_videoreference_audiowatermark

Recipes for Runway Gen-4.5

Each request below is generated from what Runway Gen-4.5 serves on NinjaChat today — the same capabilities and parameters that GET /models reports — so it runs as written with your key.

Submit and poll

Rendering takes minutes, so the API returns a job id and you poll it until the file is ready.

runway-gen4.5· submit and poll

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "runway-gen4.5",
    "prompt": "a paper boat drifting down a neon rain stream, slow dolly-in",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'
# → { "id": "JOB_ID", "status": "processing" }

curl https://www.ninjachat.ai/api/v1/videos/JOB_ID \
  -H "Authorization: Bearer $NINJACHAT_API_KEY"
# → { "id": "JOB_ID", "status": "completed", "video_url": "https://..." }

POST /videos charges the flat per-video price up front and returns a job. GET /videos/{id} reports processing, then completed with video_url, or failed — a failed render is refunded automatically.
duration is in seconds; this model accepts 4–10 seconds. aspect_ratio is 16:9 or 9:16.

Animate a still image

Start from a photo or a generated frame and let the prompt describe the motion.

runway-gen4.5· animate a still image

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "runway-gen4.5",
    "prompt": "the camera pulls back slowly as steam rises from the cup",
    "image_url": "https://example.com/espresso.jpg",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

image_url becomes the first frame. Write the prompt as motion and camera direction — the image already fixes subject and composition.
The URL must be publicly reachable; a first frame generated with POST /images/generations can be passed straight through.

Get a webhook instead of polling

runway-gen4.5· get a webhook instead of polling

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/webhooks \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://example.com/hooks/ninjachat", "events": ["video.completed", "video.failed"] }'
# → { "id": "...", "secret": "whsec_..." }   (the secret is shown once)

curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "runway-gen4.5", "prompt": "a paper boat drifting down a neon rain stream", "duration": 5 }'
# your endpoint receives POST { "type": "video.completed", "data": { "id": "...", "video_url": "..." } }

The signature is a hex HMAC-SHA256 of {timestamp}.{rawBody} with the endpoint's secret, sent as X-Ninja-Signature next to X-Ninja-Timestamp. Verify against the raw request body, not a re-serialized copy.
Deliveries retry up to 6 times, so make the handler idempotent on the job id. POST /webhooks/test sends one synthetic event to check the wiring before a real render.
You can also register the endpoint in the console under Developers → Keys → Webhooks; the same secret rule applies.

Pricing

Per video: $1.2/video

Dashboard

Runway Gen-4.5

Runway

Get an API key

Runway's Gen-4.5 text- and image-to-video model.

Modalities: textimagevideo
Price: $1.2/video
Providers: Live

Playground

Preparing playground

Providers

API

POST/api/v1/videosasync jobpoll · GET /api/v1/videos/{id}

runway-gen4.5

# 1 · start the render
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "runway-gen4.5", "prompt": "a paper boat drifting down a neon rain stream", "duration": 8, "aspect_ratio": "16:9" }'

# 2 · poll until it is ready
curl "https://www.ninjachat.ai/api/v1/videos/JOB_ID" \
  -H "Authorization: Bearer $NINJACHAT_API_KEY"

promptmodeldurationaspect_ratioimage_urlreference_imagesreference_videoreference_audiowatermark

Recipes for Runway Gen-4.5

Each request below is generated from what Runway Gen-4.5 serves on NinjaChat today — the same capabilities and parameters that GET /models reports — so it runs as written with your key.

Submit and poll

Rendering takes minutes, so the API returns a job id and you poll it until the file is ready.

runway-gen4.5· submit and poll

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "runway-gen4.5",
    "prompt": "a paper boat drifting down a neon rain stream, slow dolly-in",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'
# → { "id": "JOB_ID", "status": "processing" }

curl https://www.ninjachat.ai/api/v1/videos/JOB_ID \
  -H "Authorization: Bearer $NINJACHAT_API_KEY"
# → { "id": "JOB_ID", "status": "completed", "video_url": "https://..." }

POST /videos charges the flat per-video price up front and returns a job. GET /videos/{id} reports processing, then completed with video_url, or failed — a failed render is refunded automatically.
duration is in seconds; this model accepts 4–10 seconds. aspect_ratio is 16:9 or 9:16.

Animate a still image

Start from a photo or a generated frame and let the prompt describe the motion.

runway-gen4.5· animate a still image

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "runway-gen4.5",
    "prompt": "the camera pulls back slowly as steam rises from the cup",
    "image_url": "https://example.com/espresso.jpg",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

image_url becomes the first frame. Write the prompt as motion and camera direction — the image already fixes subject and composition.
The URL must be publicly reachable; a first frame generated with POST /images/generations can be passed straight through.

Get a webhook instead of polling

runway-gen4.5· get a webhook instead of polling

# export NINJACHAT_API_KEY="nj_sk_..."   (Developers → Keys)
curl https://www.ninjachat.ai/api/v1/webhooks \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://example.com/hooks/ninjachat", "events": ["video.completed", "video.failed"] }'
# → { "id": "...", "secret": "whsec_..." }   (the secret is shown once)

curl https://www.ninjachat.ai/api/v1/videos \
  -H "Authorization: Bearer $NINJACHAT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "runway-gen4.5", "prompt": "a paper boat drifting down a neon rain stream", "duration": 5 }'
# your endpoint receives POST { "type": "video.completed", "data": { "id": "...", "video_url": "..." } }

The signature is a hex HMAC-SHA256 of {timestamp}.{rawBody} with the endpoint's secret, sent as X-Ninja-Signature next to X-Ninja-Timestamp. Verify against the raw request body, not a re-serialized copy.
Deliveries retry up to 6 times, so make the handler idempotent on the job id. POST /webhooks/test sends one synthetic event to check the wiring before a real render.
You can also register the endpoint in the console under Developers → Keys → Webhooks; the same secret rule applies.

Pricing

Per video: $1.2/video

Runway Gen-4.5

Playground

Providers

API

Recipes for Runway Gen-4.5

Submit and poll

Animate a still image

Get a webhook instead of polling

Pricing

More from Runway

Runway Gen-4.5

Playground

Providers

API

Recipes for Runway Gen-4.5

Submit and poll

Animate a still image

Get a webhook instead of polling

Pricing

More from Runway