Advertisement|Remove ads.

Runway, an artificial intelligence startup, unveiled its latest video generation model, Gen-4.5, on Monday, claiming it as the most advanced video AI system to date.
The model promises cinematic-quality output with precise control, aiming to boost creative possibilities for content creators and enterprises alike.
Building on its previous models, including the Gen-1 launched two years ago, Runway has refined both pre- and post-training methods to optimize efficiency and output quality.
The model currently tops the Artificial Analysis Text-to-Video benchmark with 1,247 Elo points, surpassing competitors in fidelity, consistency, and controllability. According to a CNBC report, Runway’s model tops models from Google and OpenAI in terms of translating written prompts into high-definition video sequences.
Founded in 2018, Runway has grown into a recognized player in AI research and video modeling, earning a spot on CNBC’s Disruptor 50 list in 2025. Its platform serves media companies, creative studios, brands, designers, and students. PitchBook estimates the startup’s valuation at $3.55 billion.
Developed entirely on Nvidia GPUs, including Hopper and Blackwell series, Gen-4.5 leverages optimized hardware for high-speed inference without sacrificing quality.
“This is an incredibly exciting time for video and world models.”
- Jensen Huang, President and CEO, Nvidia
“Together, we are partnering to advance the entire lifecycle of AI from pretraining, to post-training and inference,” he said.
Gen 4.5 is being gradually introduced and will be accessible to all of Runway’s users by week’s end. The model will be available through Runway’s platform, API, and select partners.
Also See: Dan Ives Updates AI 30 Winner List Ahead Of 2026: SoundHound, ServiceNow, Salesforce Removed
For updates and corrections, email newsroom[at]stocktwits[dot]com.