ZMIME
Luma AI · video model · Released Feb 20, 2025

Ray2

Ray2 is a video model from Luma AI, released Feb 20, 2025. It handles text to video and image to video.

  • Video model
  • Proprietary
  • Text to video
  • Image to video
Arena score
1,108#46 of 48 · image to video
Starting price
—Not published
Longest clip
—Not published
Resolution
—No audio track
Quality & price

How it rates and what it costs

Text to video arena
1,065 ±17#46 of 48 · 5,250 votes
Image to video arena
1,108 ±16#46 of 48 · 9,521 votes

Arena scores come from blind, side-by-side human votes on the same prompts (higher is better; ± is the 95% range). Boards updated Sep 22, 2026.

Specifications

The details

Lab
Luma AI
Type
Video model
Released
Feb 20, 2025
Inputs
Text prompts, Reference images
Output
Video
Can do
Text to video, Image to video
Weights
Proprietary
Availability
1 API host
Questions

About Ray2

How good is Ray2?

In blind side-by-side votes, Ray2 scores 1,065 for text to video (#46 of 48) and 1,108 for image to video (#46 of 48). Higher is better; scores within about 34 points of each other are effectively tied.

Who makes Ray2?

Ray2 is a video model from Luma AI, released Feb 20, 2025.

What can Ray2 do?

Ray2 accepts text prompts and reference images and produces video. In practice: text to video and image to video.

Is Ray2 open source?

No. Ray2 is proprietary; you use it through an API or partner apps.

Where can I use Ray2?

It is listed by 1 API host we track.

What are the alternatives to Ray2?

The closest video models from other labs, by release date, are Veo 2 (Google, Dec 2, 2024), Seedance v1.0 Pro (ByteDance Seed, Jun 11, 2025), Runway Gen-4 Turbo (Runway, May 9, 2025) and Wan 2.2 A14B (Alibaba (Qwen), Jul 28, 2025).