Runway Gen-4 Turbo (by Runway)

Overview

Cinematic image & video generation focused on reference consistency and character continuity.

---

Quick facts

• Modes: Text + Reference Image → Image · Image → Video · Video → Video · Image → Image.

• Default output / size: 1280×720 (landscape), 720×1280 (portrait), 960×960 (square)

• Aspect ratios: 21:9, 16:9, 1:1, 4:3, 3:4, 9:16

---

What it’s great for

• Generating consistent characters from reference images.

• High-quality image-to-video and video-to-video transformations.

• Production-ready compositing and VFX workflows.

---

Example outputs

Reference character consistency — image example

Scene generation — cinematic environment

Copy-and-paste prompts

Create cinematic close-up of character from reference image, maintain facial consistency across lighting

Image-to-video: gentle camera pan around the subject, 5s, cinematic grade

Edit: replace background with rainy city street, motion-matched reflections

---

Parameters

Name	Type	Default	Notes
Prompt	Text	—	Required
Reference Image	Upload	Optional	Use for character consistency
Aspect Ratio	Select	Auto	Auto, Wide Landscape (21:9), Landscape (16:9), Square (1:1), Landscape (4:3), Portrait (3:4), Portrait (9:16)
Duration	Select	5s	5s, 10s
Seed	Seed	Random	Optional for repeatability

---

Modes

Mode	Estimated time	Required inputs	Typical use
Image → Video	1 minute	Reference Image, Prompt	Animate stills
Video → Video	4 minutes	Video, Prompt	Restyle or reanimate video
Text → Image (with reference)	45–60 seconds	Prompt, Reference Image	Character / environment generation

---

Output options

Option	Values / notes
Aspect ratio	21:9, 16:9, 1:1, 4:3, 3:4, 9:16
Notes	Reference consistency and production workflows supported

---

Prompt tips

• Include “from reference image” when asking for character consistency.

• Specify shot motion and duration for video endpoints.

---