In this episode, we finally get hands-on with Runway's newly released Gen-4 video generation model and put it through our standard dynamic motion test. Using our consistent snowboarding benchmark, we compare Gen-4 against previous Runway models (Gen-3 Alpha and Gen-3 Alpha Turbo) to measure improvements in motion handling, physics simulation, and overall realism. Through multiple tests with varying image types and prompting strategies, we discover that Gen-4 represents a significant advancement, particularly in handling natural movement and environmental interaction. While still showing some limitations with complex motions and longer sequences, the results are impressive enough to crown Runway Gen-4 as the new leader in AI video generation, especially when using high-quality source imagery and conservative prompting approaches.KeywordsRunway Gen-4AI Video GenerationMotion PhysicsDynamic MovementSnowboarding TestRunway Gen-3 AlphaRunway Gen-3 Alpha TurboReference FeatureImage-to-VideoNatural MovementEnvironmental InteractionPrompt AdherenceStock ImageryAction SequencesPhysics SimulationVideo ConsistencyModel ComparisonAI MarketingContent CreationVisual EffectsKey TakeawaysRunway Gen-4 PerformanceSignificantly improved physics handling compared to previous modelsBetter consistency throughout generated sequencesMore natural environmental interaction (snow displacement, carving)Maintains subject integrity for longer periods (solid 10-second clips)Struggles less with the geometric challenges of half-pipe structuresClear progression from Gen-3 Alpha through Gen-3 Alpha Turbo to Gen-4Our Best Practices for Gen-4Stock photography outperforms AI-generated imagery as inputConservative prompting yields more convincing resultsShorter clips (5-7 seconds) maintain higher quality throughoutSimpler motion scenarios work better than complex tricks or movementsSource images with existing environmental interaction cues improve resultsRealistic source imagery yields almost photorealistic outputsFeature HighlightsNew "reference" feature available on unlimited tier ($95/month)Allows combining multiple reference images in generationsPart of a "series" of Gen-4 models (suggesting more variants to come)Improved fidelity, dynamic motion, and controllabilityPractical ApplicationsMarketing videos requiring natural human movementSocial media content with realistic action sequencesProduct demonstrations with environmental interactionB-roll footage replacementBest for shorter clips with straightforward motionLinkshttps://runwayml.com/https://x.com/runwayml/status/1906718935778545964

Ver todos los episodios