AI Research Atlas

Gen-2

Runway · 2023-06

Gen-2 adds text-to-video and image-to-video to Runway's video-to-video tool; 8 generation modes, announced March 2023 and opened to everyone in June.

First widely available commercial text-to-video model. Users type a prompt or supply an image and get a few seconds of clip. Output was short (4 seconds), low-res and jittery by later standards, yet it made Runway the early leader before Pika, Sora and Kling.

Date
2023-06
Lab
Runway
Kind
model
Access
app only

Wikipedia lists Gen-2 as announced 2023-03 with public access in 2023-06; Runway's research page was opened but shows only 'Feb 2023' and 8 modes. Exact public-launch day not confirmed.

Sources

  1. runwayml.com/research/gen-2
  2. en.wikipedia.org/wiki/Text-to-video_model
  3. 80.lv/articles/runway-s-gen-2-has-been-officially-released/

This record was partly confirmed: some claims could not be checked on 6 October 2026. How we check

Related