Choose Auto, TokenLab Verified, or Official for each request, with prices shown up front. See what's new

Runway: Runway Gen-4.5

Runway Gen-4.5 is an AI video generation model that creates short video clips from text prompts or static images with high visual fidelity and smooth motion. It supports both text-to-video and image-to-video generation with a range of aspect ratios and clip durations. Gen-4.5 emphasizes realistic motion, strong prompt adherence, and controllable composition, making it suitable for cinematic sequences and creative video workflows.
Compare models
runway-gen-4-5
AvailableRunwayVideoAsync
Price
From $0.12
Modalities
Video

About Runway Gen-4.5

Runway Gen-4.5 is Runway's flagship video generation model, announced in December 2025 for text-to-video and image-to-video clips. Runway describes it as stronger than Gen-4 on physical accuracy, prompt adherence, and visual fidelity, at comparable speed. It targets cinematic and stylized scenes alike, from photoreal footage to animation, and sits above the Gen-4 line as the model to try first.

Where it works well

  • Object motion and physical behavior are what Runway names as the main gain over Gen-4, so falling, colliding, and flowing things look more believable.
  • Prompt adherence on busy scenes: Runway says it handles multi-element prompts better than earlier Runway models.
  • Both text-only and still-image starting points are available, so one model covers concept shots and animation of finished frames.
  • Looks range from photoreal to stylized animation without switching models.

When to choose another model

  • Runway lists causal errors, where an effect appears before its cause, and objects that vanish or appear between frames.
  • Runway notes a success bias: actions in a prompt tend to succeed even when they plausibly should fail.
  • It takes text or a single still image only; reference clips and video editing are not part of it.

Getting started

  1. Create API key

    Create a key in Console, then use it with every model on the platform.

  2. Send your first request

    Copy the example for your language and run it against the endpoint.

    Image to videoText to videoTokenLab endpoint
    POST/v1/videos/generations
    curl -X POST "https://api.tokenlab.sh/v1/videos/generations" \
      -H "Authorization: Bearer sk-xxx" \
      -H "Content-Type: application/json" \
      -d '{
      "duration": 5,
      "operation": "image-to-video",
      "model": "runway-gen-4-5",
      "prompt": "Cinematic sunrise over a calm lake with gentle camera motion.",
      "image_url": "https://example.com/image.jpg"
    }'

Parameters

Image to video

ParameterValues and limitsDefaultNotes
duration5, 8, 1010Use an integer.
height720, 832, 960, 1104, 1280720Use an integer. Use a listed width and height pair: 1280x720 (720p, 16:9), 720x1280 (720p, 9:16), 1104x832 (720p, 4:3), 832x1104 (720p, 3:4), 960x960 (720p, 1:1).
imageNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.
image_urlNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.
n11Use an integer.
operationRequiredNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 1000 characters.
seconds5, 8, 1010Use an integer.
seed≥ 0, ≤ 4294967295Not providedUse an integer.
start_imageNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.
width720, 832, 960, 1104, 12801280Use an integer. Use a listed width and height pair: 1280x720 (720p, 16:9), 720x1280 (720p, 9:16), 1104x832 (720p, 4:3), 832x1104 (720p, 3:4), 960x960 (720p, 1:1).

Text to video

ParameterValues and limitsDefaultNotes
duration5, 8, 1010Use an integer.
height720, 832, 960, 1104, 1280720Use an integer. Use a listed width and height pair: 1280x720 (720p, 16:9), 720x1280 (720p, 9:16), 1104x832 (720p, 4:3), 832x1104 (720p, 3:4), 960x960 (720p, 1:1).
n11Use an integer.
operationRequiredNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 1000 characters.
seconds5, 8, 1010Use an integer.
seed≥ 0, ≤ 4294967295Not providedUse an integer.
width720, 832, 960, 1104, 12801280Use an integer. Use a listed width and height pair: 1280x720 (720p, 16:9), 720x1280 (720p, 9:16), 1104x832 (720p, 4:3), 832x1104 (720p, 3:4), 960x960 (720p, 1:1).

Fields outside this contract are rejected.

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Image to video

per second
Official price
$0.12
Verified price
—
Discount
—

Text to video

per second
Official price
$0.12
Verified price
—
Discount
—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours
Requests
Success rate
P95 latency
Total tokens
Model performance
Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Runway Gen-4.5 in Console with a prompt ready to edit or send.

Help me create with runway-gen-4-5 using image-to-video at /v1/videos/generations. Show the result, status, and cost.

Use cases

Best for
  • Text to Video
  • Concept footage from a script

    Turn a written scene description into a short test clip to judge camera angle, lighting, and mood before any shoot or full production.

  • Animating key art

    Give it a finished still, such as a product render or poster, and describe the motion so the image becomes a short moving shot.

  • Stylized motion pieces

    Produce animated sequences for music visuals or title cards where a consistent illustrated look matters more than realism.

  • Previs for physical action

    Rough out shots involving water, cloth, or falling objects to check staging, using the model's improved physical behavior.

Prompt examples

A slow dolly-in on a rain-soaked neon street at night, a taxi passes and splashes through a puddle, shallow depth of field, 35mm look.

Animate this product photo: the glass bottle rotates a quarter turn on a marble counter while soft window light shifts across the label.

Hand-drawn watercolor animation of a paper boat drifting down a stream between mossy rocks, gentle camera pan to the right.

FAQ

What is Runway Gen-4.5?

It is Runway's video model announced in December 2025. It generates clips from a text prompt or a starting image, and Runway positions it as its most capable model for motion quality, prompt adherence, and visual fidelity, with generation speed comparable to Gen-4.

Does Gen-4.5 accept an image as input?

Yes. Runway lists image-to-video alongside text-to-video. Supply a still and a motion description to animate it, or send text alone to generate the first frame and the movement together.

How did Runway rank Gen-4.5?

Runway reports 1,247 Elo points on the Artificial Analysis Text to Video leaderboard at launch, ahead of Google Veo 3, Kling 2.5, and OpenAI Sora 2 Pro on that board. Leaderboard positions move over time.

What does Gen-4.5 still get wrong?

Runway names three weaknesses: effects occurring before their causes, objects disappearing or popping into view, and actions succeeding more often than they should. Plan retakes for shots that depend on precise cause and effect.

How is Gen-4.5 different from Gen-4?

Gen-4 focused on keeping characters, locations, and objects consistent across scenes using reference images. Gen-4.5 is the newer model with improved physics, prompt adherence, and fidelity, so choose it for new generation work unless you rely on Gen-4 behavior.

How much does Runway Gen-4.5 cost?

On TokenLab, Runway Gen-4.5 costs $0.12 per second. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Runway Gen-4.5 use?

Use https://api.tokenlab.sh/v1/videos/generations for Runway Gen-4.5. The request example below shows the matching code shape.

Which operations does Runway Gen-4.5 support?

Runway Gen-4.5 supports Image to video, Text to video. Select an operation above to see its endpoint and request example.

Compare Runway Gen-4.5

Sources

Reviewed Oct 2, 2026

More from Runway

Related models