ZMIME
Video model comparison · 2 models · Updated Nov 2, 2026

Ray2 vs Veo 2

Preferred for text to video: Veo 2 · Preferred for image to video: Veo 2 · Newest: Ray2. Quality scores are blind human-preference votes; specs and prices come from each lab’s listings.

  1. Luma AI

    Ray2

    Released Feb 20, 2025

    • Text to video
    • Image to video
    • Arena1,108
    • From—
    • Max length—
  2. Google

    Veo 2

    Released Dec 2, 2024

    • Text to video
    • Arena1,166
    • From—
    • Max length—
  3. Add a model

    Make it a three-way comparison.

01 — At a glance

Where they differ

Quality is the arena score from blind, side-by-side human votes. Price, length and resolution come from each lab’s own listings.

  • Preferred for text to videoVeo 2Ray2 1,065 (#46) · Veo 2 1,164 (#39) — blind human votes
  • Preferred for image to videoVeo 2Ray2 1,108 (#46) · Veo 2 1,166 (#44) — blind human votes
  • NewestRay2Ray2 Feb 20, 2025 · Veo 2 Dec 2, 2024
  • Widest inputsRay2Ray2: Text prompts, Reference images · Veo 2: Text prompts
  • Native audioNo native audioGenerates an audio track with the video: Ray2 no · Veo 2 no
  • Image to videoRay2Starts from an uploaded image: Ray2 yes · Veo 2 no
  • Self-hostingNo open weightsBoth are available only through APIs and apps
  • Most API hostsSame reachRay2: 1 · Veo 2: 1
02 — Side by side

Every spec in one table

Dashes mean the data is not published.

Ray2 vs Veo 2 specifications side by side
SpecificationRay2Luma AIVeo 2Google
Quality (blind human votes)
Text to video arenaElo-style score · higher is better1,065 ±17#46 of 48 · 5,250 votes1,164 ±16#39 of 48 · 6,531 votes (best)
Image to video arenaElo-style score · higher is better1,108 ±16#46 of 48 · 9,521 votes1,166 ±16#44 of 48 · 10,303 votes (best)
Output
ProducesVideoVideo
Inputs
Text promptsYesYes
Start image (image to video)YesNo
Video inputNoNo
Audio inputNoNo
Access
WeightsProprietaryProprietary
API hosts11
ReleasedFeb 20, 2025Dec 2, 2024
03 — Questions

Which should you choose?

Which is better, Ray2 or Veo 2?

For text to video, voters prefer Veo 2 (Ray2 1,065 (#46) · Veo 2 1,164 (#39)). For image to video, voters prefer Veo 2 (Ray2 1,108 (#46) · Veo 2 1,166 (#44)). Arena scores come from blind, side-by-side human votes on the same prompts; when the gap is inside both models’ error margins we call it a tie. Test your own prompts before you commit.

What is the difference between Ray2 and Veo 2?

Ray2 is a video model from Luma AI (Feb 20, 2025), taking text prompts and reference images and producing video. Veo 2 is a video model from Google (Dec 2, 2024), taking text prompts and producing video.

Which is newer?

Ray2 Feb 20, 2025; Veo 2 Dec 2, 2024.

Which generates video with sound?

Generates an audio track with the video: Ray2 no; Veo 2 no.

Can I run any of these on my own hardware?

No open weights. Both are available only through APIs and apps.