Product Launchtext to videogenerative airunway
Runway Releases Gen-4.5 Improving Text-to-Video Realism
8.0
In a Monday blog post, Runway announced Gen-4.5, an updated text-to-video model that it says produces cinematic, highly realistic outputs with improved physical accuracy and prompt adherence. The model — rolling out gradually to users — claims lifelike object motion and fluid dynamics but still shows object permanence and causal reasoning issues, raising detection and policy concerns alongside OpenAI's Sora 2 advances.
Key Points
- 1Announces Gen-4.5 text-to-video model achieving unprecedented physical accuracy and visual precision
- 2Improves prompt adherence and realistic motion, making AI outputs harder to distinguish from real footage
- 3Requires new detection, verification, and policy measures for creators, platforms, and misinformation mitigation
Scoring Rationale
Product release offers meaningful realism and official backing, but remains largely incremental amid existing text-to-video progress.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
