video model comparison
Gen-4.5 vs Veo 3.1 Preview
Gen-4.5 and Veo 3.1 Preview are both current video models, from Runway and Google. On Arena Text-to-Video, Gen-4.5 scores 1225 Elo and Veo 3.1 Preview 1364 Elo (Arena, measured September 22, 2026).
Data verified October 12, 2026 · Dataset 2026-10-12
Side by side
Standard-tier prices in effect today. The better value in each row is highlighted.
| Fact | Gen-4.5 | Veo 3.1 Preview |
|---|---|---|
| Price per video second | Not reported | $0.40 |
| Arena Text-to-Video | 1225 Elo (#4 of 4) | 1364 Elo (#2 of 4) (better) |
| Arena Image-to-Video | Not measured | 1398 Elo (#3 of 3) |
| Provider | Runway | |
| Release date | Not reported | October 15, 2025 |
| Access | API only | API only |
| Context window | Not reported | Not reported |
| Max output | Not reported | Not reported |
| Input | text, image | text, image |
| Output | video | video, audio |
Scores:Arena · 2026-10-12
When to choose each
Gen-4.5
Best for
- Runway API video generation
Watch out for
- No exact compatible public benchmark snapshot is recorded
Veo 3.1 Preview
Best for
- Gemini API video generation with audio
Watch out for
- Preview models may change and have stricter rate limits
Gen-4.5 vs Veo 3.1 Preview FAQ
Which scores higher on independent benchmarks, Gen-4.5 or Veo 3.1 Preview?
Veo 3.1 Preview leads on Arena Text-to-Video, 1364 Elo to 1225 Elo. Scores are independent evaluations, not either vendor's launch figures.
Related comparisons
Production evaluation
Choosing between Gen-4.5 and Veo 3.1 Preview?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty