Introducing our new frontier video model, Runway Gen-4.5. Previously known as Whisper Thunder (aka) David. Gen-4.5 is state-of-the-art and sets a new standard for video generation motion quality, prompt adherence and visual fidelity. Learn more below.
See What Runway Gen 4.5 Can Create
Explore a curated set of high fidelity examples generated with Runway Gen 4.5. Each clip demonstrates precise motion, consistent characters, complex scene composition and photorealistic detail trusted by creators and enterprise teams.
Core Capabilities Powered by Runway Gen 4.5
Complex and Detailed Scene Generation
Render intricate, multi element environments with precision. Gen 4.5 handles layered objects, spatial depth and dynamic lighting while maintaining clean composition and accurate positioning within a single generation.
Advanced Physical Accuracy
Produce motion grounded in natural physics. Objects display believable weight, momentum and collision behavior. Liquids, materials and surface details respond with realistic dynamics and high fidelity.
Expressive and Cinematic Character Performance
Generate characters with nuanced emotion, lifelike facial detail and natural gestures. Gen 4.5 captures subtle acting moments and delivers cinematic presence suitable for storytelling and professional content.
Photorealistic Visual Quality
Create visuals that closely resemble real world footage. Fine details such as hair strands, skin texture and material patterns remain coherent across motion resulting in highly convincing output.
Versatile Non Photorealistic and Artistic Styles
Explore stylized looks unconstrained by realism including stop motion aesthetics, painterly rendering and fantasy animation. Gen 4.5 adapts fluently to creative direction across diverse visual styles.
Optimized Performance on NVIDIA Hardware
Runway Gen 4.5 is built and optimized on NVIDIA Hopper and Blackwell GPUs. The model benefits from end to end collaboration with NVIDIA across training and inference providing production grade performance at scale.
What Is Runway Gen 4.5
Runway Gen 4.5 is the newest generation text to video model released on December 1, 2025. It represents Runway’s most advanced system to date with major improvements in accuracy and overall generation quality.
In public benchmark testing Gen 4.5 achieved an Elo score of 1247 ranking first on global text to video leaderboards. It provides creators and enterprises early access to one of the highest performing video generation models currently available.
Who Benefits from Runway Gen 4.5
Solo Creators
Turn ideas into polished video without cameras or crews. Create scenes up to 90% faster so your inspiration never waits for production.
Creative Teams
Move from concept to usable footage in minutes. Iterate freely and cut preproduction time by nearly 70 % while keeping visual quality high.
Marketing Pros
Test multiple campaign directions in a single afternoon. Reduce briefing and revision cycles by up to 60 % with precise prompt to video alignment.
Brands and Enterprises
Produce product showcases and storytelling content at scale. Lower content costs significantly while maintaining consistent visual standards across all assets.
Filmmakers
Preview complex shots, performances and lighting setups before stepping on set. Improve decision making speed by more than 50 % during planning.
How to Create AI Videos in 3 Simple Steps
01
Describe Your Vision
Type a text prompt describing your scene, camera movement, lighting, and mood. Be as detailed as you want Gen 4.5 understands complex instructions.
02
Generate & Preview
Watch AI bring your description to life in under a minute. Generate multiple variations and compare them side by side. Choose the perfect take.
03
Export & Share
Download in high definition up to 4K. Drop it into Premiere, Final Cut, DaVinci wherever you edit. Or share directly to social platforms.
Why Creators Can’t Stop Talking About Runway Gen 4.5
Video Gen progress is amazing 🔥 Hearing great things about Runway Gen 4.5.
🎆 Video Model Comparison: Text to video There are a few new models in this comparison, including Wan 2.5, Kling 2.5 & Luma Ray3: → Wan 2.5 → Kling 2.5 → Luma Ray3 → Veo 3 → Seedance Pro → Hailuo-02 → Runway Gen-4 → PixVerse 5 I used the same prompt on each model up to
"Can Runway's new model handle more traditional 2d animation styles?" Yep 😎 Gen-4.5, also known as Whisper Thunder aka David, will be gradually rolling out to everyone in the coming days.
For nostalgia’s sake, here are the same prompts run on the world’s first available text-to-video model in 2023 ( @runwayml Gen-2) and the world’s best current model (Runway Gen-4.5)
i just got access to runway gen-4 & it's insane... i'll add the things i generate to this thread 👇
Frequently Asked Questions
How is Gen 4.5 different from other AI video generators like Sora or Veo?
Gen 4.5 ranks #1 on the Artificial Analysis Text to Video Leaderboard, outperforming Google Veo 3, OpenAI Sora, and all other models in blind human evaluation. It delivers superior motion quality, prompt adherence, and visual consistency.
Do I need technical skills or video editing experience?
No. Runway is designed for creators of all skill levels. Describe what you want in plain language, and the AI generates your video. No coding or editing experience required.
How long are the generated videos?
Gen 4.5 creates video clips of 5 10 seconds per generation. You can extend videos using Runway's expansion tools and combine multiple clips in your editing workflow.
What output resolutions are available?
Generate in standard definition and upscale to 4K for crystal clear, broadcast quality video output.
What generation modes does Gen 4.5 support?
Gen 4.5 supports text to video, image to video, video to video transformation, keyframe control, and more, giving you multiple ways to bring any creative vision to life.
Start Creating with Runway Gen 4.5
Experience the newest generation of AI video creation with access to Runway Gen 4.5.