Introducing RunwayML Character Performance, Video-to-Video, and Video Upscale in Griptape Nodes

Creative professionals are entering a golden age of generative AI, where the ability to rapidly prototype, iterate, and deliver stunning visual content is transforming industries. With Griptape Nodes, we’re committed to keeping you at the cutting edge. Today, we’re thrilled to introduce a trio of powerful new features for our RunwayML integration that will supercharge your creative workflows.

Animate Characters Like Never Before: RunwayML Character Performance

Imagine breathing life into a static character image or video, making it perform actions based on a reference video, no mocap suit required. With the new RunwayML Character Performance node, powered by RunwayML’s Act Two model, you can animate characters with nuanced body control and expressive intensity. Simply provide a character image or video, a reference performance, and let the node handle the rest. Whether you’re prototyping game animations, crafting cinematic sequences, or experimenting with digital puppetry, this tool unlocks a new dimension of creative freedom. For the best results and seamless video transcoding, we recommend installing ffmpeg on the system where you are running your Griptape Nodes Engine.

In the example below I have a studio clip of a presenter talking to the camera and I have coupled this with a generated image of a cartoon character in a snowy alpine environment.

Transform Video with Text Prompts: RunwayML Video-to-Video

What if you could take an existing video and reimagine it, guided by nothing more than a text prompt? The RunwayML Video-to-Video node makes this possible, using RunwayML’s advanced generative models to reinterpret your footage in entirely new ways. Feed in a video and a descriptive prompt, and see your content morph to match your creative vision. This is a game-changer for advertising creatives looking to iterate on campaign assets, VFX artists seeking rapid concept development, and anyone eager to push the boundaries of video storytelling. Again, ffmpeg is recommended for optimal transcoding.

For this example I generated a 960x960 video of an alpine scene. To generate this video, I started by using the Runway Text-to-Image Node with the following prompt:

A breathtaking aerial view over the Swiss Alps in summer, showcasing vibrant meadows bursting with colorful wildflowers, crystal-clear mountain lakes reflecting the blue sky, and charming ancient wooden chalets nestled among lush green hills, all under a bright, sunny sky with soft, fluffy clouds.

I then combined this with the Runway Image-to-Video Node with the following prompt to generate a 10 second video from the image:

Shot from a drone. The drone gains altitude and flies towards the lake.

For the Video-to-Video Node, I combined the 960x960 output video with the original static image as a reference image, and added the text prompt:

Convert this to a winter scene and add several colorful hot air balloons

You can see the results in the video below, which shows both the input video and the output. Using the original image a reference helps to ensure scene consistency, which I think works well in this example. 

Crystal-Clear Upscaling: RunwayML Video Upscale

Resolution constraints are a thing of the past. The new RunwayML Video Upscale node lets you boost your video’s resolution by up to 4x, capped at a stunning 4096px per side. Whether you’re preparing assets for high-end displays, refining game cinematics, or simply want your work to shine in ultra-HD, this feature ensures your visuals are crisp, detailed, and ready for any platform.

For this example, I used the same 960x960 video of an alpine scene and I then used the Runway Video Upscale node to upscale this video to 3840 x 3840 and got the results you can see in the video below.

Why Griptape Nodes?

Griptape Nodes is more than just a workflow engine. It’s your creative playground. With drag-and-drop simplicity, seamless API integrations, and a growing library of AI-powered nodes, you can build, automate, and share complex creative workflows without writing a single line of code. The latest RunwayML features are designed to slot effortlessly into your existing pipelines, empowering you to experiment, iterate, and deliver at the speed of inspiration.

Ready to Elevate Your Creative Process?

Don’t let your creative potential be limited by technical barriers. Sign up for Griptape Nodes today and unlock the full power of generative AI in your projects. Explore the new RunwayML Character Performance, Video-to-Video, and Video Upscale nodes, available now in the Griptape Nodes library for RunwayML. Visit www.griptapenodes.com to get started, access detailed setup guides, and join a community of forward-thinking creators on the Griptape Discord.

What will you create next? Dive in, experiment, and share your results. We can’t wait to see how you push the boundaries of what’s possible with Griptape Nodes and RunwayML.

Try Griptape Nodes now and transform your creative workflow.